TW200916986A - High resolution display - Google Patents

High resolution display Download PDF

Info

Publication number
TW200916986A
TW200916986A TW97118290A TW97118290A TW200916986A TW 200916986 A TW200916986 A TW 200916986A TW 97118290 A TW97118290 A TW 97118290A TW 97118290 A TW97118290 A TW 97118290A TW 200916986 A TW200916986 A TW 200916986A
Authority
TW
Taiwan
Prior art keywords
display
image
holographic
resolution
data
Prior art date
Application number
TW97118290A
Other languages
Chinese (zh)
Other versions
TWI464548B (en
Inventor
Robert Mibbach
Alexander Schwedtner
Bo Kroll
Original Assignee
Seereal Technologies Sa
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from GBGB0718602.6A external-priority patent/GB0718602D0/en
Application filed by Seereal Technologies Sa filed Critical Seereal Technologies Sa
Publication of TW200916986A publication Critical patent/TW200916986A/en
Application granted granted Critical
Publication of TWI464548B publication Critical patent/TWI464548B/en

Links

Landscapes

  • Holo Graphy (AREA)

Abstract

Disclosed is a high resolution display on which decompressed high resolution image data is displayed, the display including pixels, the pixels being on a substrate, where circuitry is present on the same substrate as the pixels, where compressed high resolution image data which has been compressed using known data compression techniques is received by the circuitry, the circuitry being operable to perform decompression calculations which provide subsequent display of decompressed high resolution image data by the pixels of the display.

Description

200916986 九、發明說明： 【發明所屬之技術領域】 —本舍縣與-個可齡經過解壓辆高解析度影像龍的高解析度顯 不波置有關’顯不ϋ她綠像素，這些像素位於—織板上。這個高解 ，度顯示裝置同時可顯示—般的影像資料；這個高解析度顯示裝置也可以 是可顯示全像顯示器資料的顯示器。 【先前技術】 電月劫產生全像影像（CGH)會在一個或多個空間光調制器⑽）中進 行編碼處理；SLM可能包括許多可透過電學或光學方式控伽像素格。这 些像素格會__應於—舰齡像影像的全像影像魏行編碼來調制 光線的振幅及/或相位。CGH可以透過計算得出，例如，可透過‘‘―致光 線追縱”、透賴擬由景祕反射光_—個參考光波之間造成的干涉、 或者透過傅立葉（Fwier)或菲涅爾（Fresnei)轉換來進行計算。一個 理想的Μ將可嶋㈢嫩錄值，⑽卩，可物卿—個輸入 的光波的振幅及她。不過…個典動SLM則只會控制—個性質，不是 振幅就是相位，而且還會有也可能會影響另—個性質不良副作用。有許^ 不同的方法可以用來依振幅或相⑽光線進行調制，例如，有電學編址二 液晶SLM、光學編址式液晶SLM、磁光學式⑽、微反射魏置:工 調制器等。光的調制可以是空壯連續的或者包括有多個m 式排列、可個別編址（二進位，多階或連續）的像素格。 、 在本谢，所謂的“編碼”表示對—個魄物的各個區域供 200916986 應各個控制值來編寫成一個全像影像使它可以透過SLM來重建出一個犯 景象的方法。 和純粹的自動立體顯示器不同的是，透過視訊全像影像，一個觀察者 可以看到-個二維景象的光波前的光學重建。31)景象會重建在一個在觀察 者的眼睛與空間光調制器⑽）之間展開的空間之中。SLM也可以以視訊 王像衫像進彳了編碼使觀察者可以在SLM的前方看到重建的三維景象的物 件，而以及在SLM上或SLM背後的其他物件。 空間光調制器的像素格最好是可讓光線通過的可透光像素格，它的光 線可以在至少-個定義的位置上以及以—致的長度（幾絲左幻產生干 涉。這可讓全像影像鍵至少在_個維度上具有__#的解析度。這樣 的光線我們將它稱為“充分一致的光線”。 為了在時間上確保具有足夠的一致性，由光源所發出的光的光譜必須 在-個適當狹㈣波長範_，也就是說，它必須接近單色I高亮度發 光-極體（LED)所發出的光的光譜帶寬足触窄，可以確保全像影像重建 在時間上的-致性。在SLM處的繞射肢會與波長成正比，這表示只有單 色的光源會產生制的物件‘_重建。較寬的光譜將會導致較粗大的物件 點而會使«的物件_不清。—個雷射絲的光射以被視為單色光。 LED的光線在空間上的線寬足夠狹窄，可以促進良好的重建。 工間上的-致性與光源的側向寬幅有關。傳統的光源，像是咖或a 陰極營光燈⑽L)，如果_—個____線，則也 能符合這些縣。—㈣射光源所糾的光可靖它視諸—個在繞射限 200916986 織圍之内的點狀光源所發出的光，並且，根據它“形態上的純淨度”， 可乂產生糊的物件重建’也就是說，每—個物件轉會被鍵為—個在 繞射限制範圍之内的點。 攸個在工間上不一致的光源所發出的光會有較大的側向寬幅而造成 重建的物件模糊不清。這個模_程度是因為在—個特定的位置所重建的 .-個物件關大小加寬職。為了可贿用在空間上部-致的光源來進行 全像衫像的重建，必須透過—個透光孔在亮度與關絲的側向寬幅之間 找出-個*協點。光源越小，它在空間上的—致性越佳。 -個直線光源，如果從—個與它的縱長直角相交的角度來看，也可以 視為是-個點狀光源。在這個方向上，光波可以一致的傳播，但是在另— 個方向上則會不一致。 整體而言，-個全像影像會透過光波在水平及垂直方向上的一雌重 4來以全像方式鍵-個景雜的—個視齡像影㈣為“全視差全 像影像”。它所重建的物件可峨水平及垂直方向看並且具有運動視差， 就像是-個真實的物體-樣。不過，如果要具有寬大的視角，則必須在μ 的水平及垂直方向上都具有極高的解析度。 通系，對於SLM的要求會因為將它限制為一個“唯水平視差，，（Ηρο) 的全像影像喊少；它只會在水付向上進行全像影像重建，而在垂直方 向上則不會進行全像影像重建。這會導致所重建的物件只具有水平運動視 差而在垂直運動時匕的透視景觀則不會變化。一個Ηρο全像影像在垂直 7 200916986 方向上對於SLM的解析度要求比—個全視差全像影像低。“唯垂直視差” (VP0)的全像影像也是可行的，但並不常見；這種全像影像重建只會發生 在垂直方向上，因而使所重建的物件只具有垂直運動視差，而在水平方向 上則不會有運動視差。因此，左眼和右眼所見的不同透視景觀也必須分別 建立。 全像影像的即時計算需要非常強大的運算性能，目前這可以透過昂貴 且特別製造的硬體設備來實現，例如具備元件可編程邏__ (FpGA)、 π王各製化1C、或是針對應用指定的碰電路⑽G)的設備，或者使用 可以平行處理的多個中央處理單元（GPU)來計算。 在薄膜電晶體（TFT)顯示器上，互相垂直的方向上的像素間距會決定 每個像2的面積。廷個面積被劃分為用來進行液晶⑽控制的透明電極、 _ M及縱列和橫顺路。騎縱躲路的線數鮮要求和顯 尺寸會决疋所要求的外形，因此也會決定橫列和縱列線路的寬度。 2的全输嶋有比目前市面上所供應請式監視器裝置 ==鮮伽物物細^物嫩絲個像素面積 辦。㈣糊獅__肢w孩比例的大 二二透明電極可用的面積大幅減少，而使得顯示器的透光度顯著 限制下=:==^_頻率的高解析度全像顯示器只有在嚴格的 行全像= 於對運算性能的要求極高，目前可以用來進 象即時計⑽軸，賴《物-峨的硬體設備， 200916986 、叩貝。由於它牽涉到大量的數據資料，從運算裝置到顯示 器的影像資料傳輸同樣也非常困難。 以下將參考先前的專利文獻資料(請參_ 1G，取自美國US6，153,893 號專利），針對—種主動矩陣式液晶顯示裝置的共通構造加以簡要解說；本 文件中引用US6,153, 893號專利僅做為參考用途。如圖1〇中所示，這個 主動矩陣編㈣—卿板物造，嫩-似板⑻、-片 相對基板脱、和—個將主基板與相對基板固定在-起的間隔空間103, 而液晶職爾挪紙輪㈣馳綱表面上構成 一個顯示部份⑽，這個部份_像素電極iq4和絲_列成矩陣 的像素電極104的開/關奘署 ^ 、以及連接到顯示部份106上的週邊 _部份心開/關裝置邮即由薄膜電晶體所構成；而薄膜電晶體同 時也包含在週邊部份107中構成電路元件。 由本申請人所提出的第W0 2刚/〇_6號文件（合併於本文件中做 ^考），其中制有—射崎計算電腦產生全像影像的方法。根據這個 方法’具㈠目三峨的·趣糊物輪分啸個平行的虛 擬剖面層的點矩陣中使每—個剖面層都有—個以點矩陣中的不連續振幅值 所疋義_物件的資料集，而針對_個全像影像__光_ 所進行的讀全像編侧__影_觸料算得出。 。 根據本申請人所提出的第奶繼/〇25839號文件（合併於本文件中 做為參考），下列步驟將透過_的協助來執行： '針對一個位於有限的距離處 处與各剖面層平行的觀察者伟，根據每— 200916986 個斷層景 計算出高布的形式 中的至少〜個_ 〃 a針對位於接近—個觀察者雙眼的觀察者平面 者視窗的虛擬觀察者視窗計算4所有剖_波場，前述的觀察 面積相較於視訊全像影像較為縮小； 面有關的 加入電腦計算出來的所有勤層的分布來針料觀 個麵中術侧梅—顧細；…’ 像資料集轉換為—個位於—個有限距離處且與參考平面平行的全 像:料率ιΓ針對一個景象的匯集電腦產生全像影像建立—個全像影 後在觀_梅繼設於全細_中，且射_於編碼 後在觀察者眼睛前方的空間中透過前述空間光調制器的協助進行重建。 ，上面提到的方法及顯示器都以不是重建景象物件本身、而是在一㈣ =個波前將由物件本身所發出的虛擬觀察者視窗中進行重建的構想為基 觀察者可以細Μ齡姻輪她。趣嶋視窗备覆蓋 觀察者眼睛的瞳孔並且可以透過已知的位置_及追縱錢的幫助來追縱 貫際的絲者位置…健_、_賴職的重建郎會展開在全像 影像顯示跑咖與嶋視㈣，其t su ^這個截頭錐 體的底面，而觀察者視窗代表它的頂面。如果觀察者視窗非常小，則這個 截頭錐體可能會接近-個金字塔形狀。觀察者透過虛擬觀察者視窗朝向顯 不器的方峨看，並在觀察者接收代表景象的波前。祕必要的轉 換的量非常大，全像編碼的過程會造成龐大的運算負荷。即時編碼也將會 200916986 需要飛常昂貴的高性能運算裝置。 本申請人所提出的第w〇細8/〇2583 «具有緩衝„輯三維影料即時產個可以讓人們 得利用刪彻崎算n剩方法。這使 本申健以生_讀影像的作法成為可能。 *休月人馳出的第W〇 2〇_2_號文件揭露了-個可即時產生 电腦產生全像練財法。透 ^ 建構__# / 個工間先调制為則上的各個物件點所 ==t _啸辑_«峨_資訊的 〜像貝料進偶碼。與上述先前 簡/025839號文件中所揭m •决方法類似，第W0 /… 揭路的方法同樣以不是重建景象物件本身、而是 前將由物件本身所發出的虛擬觀_ ;先為=個經過調制的波場會透過—個空_制器μ從充分一致 的謝產^峨蝴彡输齡 三維景象則透過在空間中所發"際次虛擬 Slii滅-滅 發生的干涉未進仃重建。虛擬觀察者視窗會以 '、、、4在各個截頭錐體形狀的重建空間中產生。這歧視窗會落 在接近觀察者目w的位置，並且可以透配知的位置細 助來追蹤實際的邀宛去办里咕 啊示、死的m 所«的是π==ΓΜ5δ39號文件中所揭露的方法 贿者看見一個景象的這個區域是由-個從SLM 一直伸展_______ 錐體大致塊^目綱職，嫩繼輸㈣= =Γ個方法也以一個單—物件點的重建只需要—個次全像影像做為 的一個子麵糊為基礎。因此與每—個景象时_:#訊並不會分 200916986 布到整個全像影像中，而只會包 全像影像，，中。依照這個概念，景象中_疋&域内，也就疋所謂的次 6λ個1斿主 ^中的—個個別物件點只會透過SLM上 的-個限疋的像素區域進行 續_號文件所揭露的觀點是以2的二欠全像影像”。第奶 的重建的每-個物件點可峨 时全像影像《於整個景象 似細視表擷取、而這些次全像 以便針對整個景象的重建構成— 累積 根據第基礎。 範例，㈣崎 定。每一個觀_分配㈣察者向來決 的位置處的-倾察者平面中 碰觀祭者眼睛 仕個預備處理步驟中，| 一 分解而成為許許多多的可見物件點。這些資料也 自1仃二維 W〇2〇_2_號文件中所揭露的處理步驟包括： 個介面。第 步驟 像的位置和範圍 y座標和它的缓 找出每-個物件點社全像影像驗[相觸次全像影 會根據-個物件點的位置導出，也就是根據㈣橫向/ 衝區距離。 —步驟2 : 從檢視表擷取相應的次全像影像的配置。 一步驟3 : 加這些次全像影像以便 針對所有的物件點重複進行上述這兩個步驟，並累 針對整個景象的重建構成—個總體的全像影像。 12 200916986 根據第W0 2_/025839號文件中所揭露的一個簡單範例，指定給一 個物件關-個次全縣像的大小會根據奴'較理糾。錢瞳孔的觀 察者視窗或它的-雜會經由物件點投射到全像影像平㈣，也就是投射 到⑽上。並據此來決定重建這個景象點所需的次全像影像的像素指標。 根據第WG 20關25839號文件所揭露的—個進—步的觀點，將會對 各個次全像f彡像或總體全像影像施加瓣祕正魏，例如，聽補償因 它的位置to狀所造成的SLM的誤差、或者用來改善重建的品質。舉例來 說’將修正值加入到次全像影像及/或總體全像影像的資料值中。另外， 由於每-個次全像影像都是由觀察者視f的實際位置所定義的，因此可以 針對較不平常的觀察者(例如，如果觀察者從旁_位置以一個極大 的角度觀看顯示器）產生一些特殊的檢視表。 如第W0 2008/025839號文件所述，使用檢視表的原則還可以根據喜 好來擴大。例如，麟色彩和亮參數:雜就可_存在單獨的檢 _中。糾’次全像影像及/或總體全像影像的資料值也可以配合來自 檢視表的亮度及/或色彩值來進行調制。其中的色彩表現是以各個原色可 以從個別的檢視表擷取的構想為基礎。 第W0 2008/025839號文件中所揭露的方法所根據的触表偏好依照 第W0 2006/066906號或第W0 2006/_19號專利文件（由本申請人所 提出，合本文射做為參考）產生。之後並鎌視表儲存於合適的資 料載體及儲存媒體中。 露的以只有一位觀察者為 圖26A說明第TO 2008/025839號文件所揭 13 200916986 例的一般構想。-個景象⑻的視域會根據—個觀察者⑼的位置和觀 看的方_定。__被分配_—健_察_⑽）， 而德窗麻椒嶋_的―吻樹。—無過調制的波 場會透過—個簡咖⑽）心-峨、魄，細各個全 像影像值來控制。這個方法和透過這個方法所導出的顯示器則以並不重建 景象的物件本身崎在—個或多個虛擬觀察_()中重建物件所發 出的波前的構想為基礎。在圖26A中，物件以一個單一物件點（pp)絲 p觀察者⑼可以透過虛擬觀察者視窗（簡）而看到景象⑻。虛擬 觀察者«⑽）會覆_者⑼_瞳孔，並且可以在已知的位 置谓測及追«、統的協助下追轉__者位置。藉倾訊全像影像的 全像影像值來控制空_制器⑽）並據此造成波場（這會在各娜素 I進仃_亚從顯示器螢幕發射出來），透過在重建空間中產生干涉來重建 三維景象。如關中所可見的，根據這個顯示設計的—般原則，景象 (s)的一個單一物件點（pp)只會由空間光調制器（⑽）上的一個限定 的像素區域重建’即所謂的次全像影像⑽。如同圖脱令所可見的，根 =個締單的解決方案’一個次全像影像（SH)的大小會根據交叉線定 ^來决定’之後再藉此得出重建這個物件點（〇p)所需的像素的指標。次 全像影像（SH)的位置和範圍則根據一個物件點（pp)的位置（也就是它 的㈣X、y轴座標）和它的緩衝區距離（也就是Z軸距離）導出。然後， 用來重建逆個點（pp)所需的個全像影像值則可以從檢視表（咖）擷取。 人全像影像（SH)會以-個亮度及/或色彩值來進行調制，然後累加 14 200916986 到個別位置處的全像影像平面中以便構成—個所謂的“總體全像影像”。 包含在上述檢視表中的資料會事先產生。這些資料偏好採用第w〇 2006/066906號文件中所說明的方法產生（如前面所節錄的先前的專利中 所述），並儲存在合適的資體及齡媒體巾。透過物件關位置和性質 的幫助，可以事先計算出各個相應的次全像影像以及次全像影像的檢視 表，並可據此產生色彩和亮度值以及修正參數。 圖26B則更詳細的解說了這個原理，並顯示出分別被指定給各個物件 點m、P2)的次全像影像⑽、SH2)。在圖26β中可以看到這些次全像 影像會被限制並構成總體全像影像，也就是整個空間光調 (slm)，的/ -個小而連續的子集。除了根據交叉線定理來定義次全像影像的位置和範 圍之外，如圖26中所可見的，還可能會有其他的函數關係。 3.探討先前的專利 由本申請人所提出的第WO 2__59號⑽咖/_994)及第 觀漏·號專利（全部合併在本文件巾做為參考）說明了—個透過充 分-致的光線·射來重建三維景象的裝置；這個裝置包括有—個點狀光 源或直線光源、-_來進行光線聚焦的透鏡、和—個"光調制器 _。和傳統全像顯示器不一樣的是，這個SIJ在傳輸模式中會在至少 -個“虛擬觀察者視窗”（關於這個用語以及相關的技術的討論，請參閱 附錄I及II)中重建-個3D景象。每一個虛擬觀察者視窗會落在靠近 觀察者眼睛的位置並且會限制在—定的大小使虛擬觀察者視窗落在一個單 -繞射級中’讓每個眼睛都能在—個從Μ表面伸展到虛擬觀察者視窗的 15 200916986 截頭錐體雜的重建空財相完整的三維景象重建。為了讓—個全像影 像重建完全沒錢動，虛峨察者視㈣大小必财超過—個重建的繞射 級的週期間隔。不過’它也必須至少大到足以讓—位觀看者可以透過這此 視窗看到整個重建的3D景象。另一隻眼睛可以透過同—個虛擬觀察者視 窗重建的景象，或者可以指定第二個虛擬觀察者視t (由—個第二光源根 據相關計算產生）。在這裡…個能見區域（這通常會相當大）它的位置會 被限制在虛峨察者視窗上。6知的解決絲會則、巧的方式重建從一個 傳統的SLM表面的高解析度所產生的大面積，織在將它縮小到虛擬觀察 者視窗的大小。由於幾何上的因素，這會有繞射角度極小的效果，使得現 行世代的SLM的解析度即足以透過合理的消費等級的運算設備而達到一 個品質極高的即時全像影像重建。 /第麵4/G22_物|](全部合併在本文件中做為參考 路了-個可產生三維立體影像的行動電話4過，在 維立體影像是利用自動奸日μ 路的二 生的：维立以自動讀助方式所產 的-、准讀4的-個問題是：通常歸麵 器的内側，而觀看者眼睛的的焦點則通常會有聚焦在顯二 這個觀看者眼睛的隹點谕仏汗> ,l 丁益表面的傾向。 在許多雜W 到維立體影像的位置之_不-致， 全像攝影技_生三維續影賴航下，這==覺。在以 以顯著的減少。 會發生、或者可 【發明内容】 資料的高解析度顯 本發明提-射顯祕過解壓賴高解析度影像 200916986 不器，顯示it包括無數像素，這些像素位於—個基板上，其中的電路組件 ^於與各自如；其巾，、晴_纖度影像 爾用已__技舰行齡蝴輪峨收；電路組 <的作用了執订解壓縮計算，並提供經過解壓縮的高解析度影像資料，由 顯不器上的各個像素進行後續的顯示。 根據上频想，射鶴舰由佈置在躺轉的像素所在 的同一基板上的電路組件執行。 根據上輯想，射的解珊算會·分布在遍及整觸示器像素基 板上的電馳件以-個分散的方式執行。 _上麵想，射，以彻軟_行_程紅經改為利 刀放在整個顯不器像素基板上的電路組件在硬體中執行。 根據上榻H巾侧_爾雜@ 器的各個像素1。 ，置在‘，、、員不 _上述構想，其中可以用來執行解壓縮計算的電路組件佈置在包含 柄的各轉她㈣，蝴齡爾細_—基板上。 根據上讀想’其巾經過__料會傳送咖示器 =都咖咖每嶋梅對所= 縮的資料Γ自轉峨，嶋糊細梅編經過賴 根據上述顯’其巾的顯轉可靖示—般顯示資料。 據上述構想’其中的顯示ϋ可以顯示全像顯示資料。 17 200916986 根據上述構想，其中用來執行壓縮計算的空間係與顯示器的各個像素 位於同一基板上。 〃 像素在 根據上述構想，其中时執行壓縮計算的空間與顯示器的各個 不同一基板上。 的橫列 根據上述構想，其中用於進行解壓縮計算的叢集會經由顯示器 及縱列線路接收資料。 資料 根據上述構想’其中每個用於解壓縮計算的叢集會經由—個並列 匯流排接收資料。 根據上述構想，其巾每細於賴縮計細叢齡軸— 連線接收資料。 貝枓 根據上述構想，其係為一個極高解析度的顯示器 根據上述構想，其中的解壓縮會由每個叢集在 執行 40 ms或以下的時間内 根據上述構想，其中的全像影像計算會在解壓縮之後執行。 執行 根據上述構想，其中用來決定—個空間光_器的編碼所執行的叶算 會利用魅在與空間光湘㈣各個像麵在_—基板上的電路組件^ 根據上述構想，其中用來決定一個空間光調制器的編碼所執行的叶算 並不牽涉到傅立f轉換（Fourier transfQrm)麵涅爾轉換如咖异 transform)的計算。 Snel 根據上述構想，其中用來蚊-個空間光調制㈣編碼所執行的叶算 200916986 會利用㈣纖__各繼之_電馳件來執行。 根據上述，射絲決定-财的編觸 進行各個 會在分散在顯示”的各爐域執行，以便針對各個分散區域，计异 相應的分散區域的像素的編碼。 根據上述構想， 根據上述構想， 構成。 其中的電路組件包括有薄膜電晶體。 其中至少有-部份的電路組件的有效區位由多晶石夕所 根據上述構想，其中至少有一部份的電路組件的有效區位由連續晶粒 石夕所構成。 根據上述構想，其中至少有—部份的電路组件的有效區位由多晶石夕錯 所構成。 根據上述構想’其中至少有—部份的電路組件的有效區位由單晶石夕所 構成。 根據上述構想，其巾至少有—部份的電輪件的有效區位由單體晶粒 石夕所構成。 根據上述構想’其中至少有—部份的電路組件的有效區位由有機半導 體所構成。 根據上述構想，其中的基板採用單晶矽。 根據上述構想’其中_示器在製造上採时晶液晶技術。 根據上述構想’其中的基板採用玻璃。 根據上述構想，其中的視訊影格速率至少約為25 Hz。 19 200916986 根據上迷構想，有真實^ I- 'f e 、像資枓會被傳輸給顯示器。 根據上迷構想，其中的真實空 構成。 綠歧縱W應資料所 根據上述構想’射⑽Wg卩輪即時運算。 ==，其中所進行的全像計算會_視表方法來執行。 根據上顿想，財_:欠_縣斯全像計算。 根據上述構想’射的_在製造上採用_技術。 根據上述構想，其中顯示器在製造上將採晴射顯示靡）技術, 本發蚊提出-街_所述㈣想之高解析度齡器來顯矛 、踢解壓_高解财影«料的方法，包括下列步驟： (a) 壓縮高解析度影像資料； (b) 傳輸經過磨縮的高解析度影像資料； (c)接收經過壓縮的高解析度影像資料； ⑷對經碰_高解析度輯㈣進行麵縮；以及 (e)將經過解壓縮的高解析度影像資料顯示在顯示器上。 本發明可藉由下狀較佳實蘭加以說明： 圖1是用來顯示全像影像的資料傳輸速率遠高於原始真實空間資料的 資料傳輸速率的說明圖； 圖2疋先前專利中的SLM的-部份與一個可以在像素矩陣的空間中 行王像4算的SLM的-部份在構造和性能特性方面的比較圖； 圖3是一個可以在像素矩陣的空間中執行全像計算的SLM的一部份 200916986 的構造圖； 疋個可以在像素矩陣的空間中執行解壓縮計算以用於全像資料 顯示的SLM的—部份的構造圖； 圖5疋個可以在像素矩陣的空間中執行解壓縮計算以用於傳統邪 顯示資料顯示的SLM的—部份的構造圖； 圖6是顯示TFT的製造過程中的各個情況的說明圖； 圖疋.‘、員不TFT的製造過程中的各個情況的說明圖； 圖8是顯示根據本發明的一個顯示設計來重建全像影像的方法的說明 圖， 圖9是顯示根據本發明的—個顯示設計來重建全像影像的方法的說明 圖； 圖10疋以先别的專利為根據的—個傳統絲矩陣式液晶顯示裝置的 一般構造的透視圖； 圖11包括顯示以本發明的—個顯示設計為根據的全像顯示涵一個 主動矩陣式基板的各個製造步驟的說明圖； 圖已括肩示圖11中的主動矩陣式基板的各個進一步製造步驟的說 明圖； 圖13包括顯示圖12中的主動矩陣式基板的各個進一步製造步驟的說 明圖； 圓1疋個在各個刀政且任意的位置上表現各個物件點的全像顯示 器的說明圖； 21 200916986 圖15是可能提供在以本發明的一個顯示設計為根據的全像蔡厂。。 的圖形計算中的功能單元的說明圖； 圖16是用於以本發明的一個顯示設計為根據的全像顯示器中的次入 像影像SH的檢視表的說明圖； 圖17是用於以本發明的-個顯示設計為根據的全像顯示器中進行全 像轉換與編碼的額外處理單元的說圖； 圖18是顯示在以本發明的一個顯示設計為根據的全像顯示器中，如果 知用次全像影像’運算負荷將會更小（因為像素格的數量更少）的說明圖. 圖19是顯示-個在時間"寺顯示的景象、另一個在時間讯時顯示 的景象、以及兩者之間的景象差別的說明圖； 圖20是齡具備有可編址㈣傳輸能力的以本發_—個顯示設計 為根據的全像顯示裝置的說明圖； 圖21顯不-個會在其中計算以本發明的_個顯示設計為根據的全像 顯示器中的電晶體數量的試算表的一部份； 圖22是圖21中所示的試算表的其餘部份； 圖23是以她剛全繼綱根據 設計的簡圖； ” 欠圖24疋乂用於本i明的一個顯不設計的全像顯示裝置為根據的顯示 資料所採取的路徑的說明圖； 〜圖25是項顯示的運算上的各項計算可能在像素矩陣的空間中 可以‘.、，員不傳統2D顯不資料、或全像顯示資料的一個灿的—部 22 200916986 份的構造圖； 圖26疋以先前的專利為根據、用來產生次全像影像的方法的說明圖； 圖7疋頌示根據本發明的一個顯示設計來重建全像影像的方法的說 明圖； 圖28疋以本發明的一個顯示設計為根據的面板拼貼的構造圖； 圖29是關於‘‘吸收”的幾何上的考量的的說明圖； 圖3〇是_ “吸收，’的幾何上的考量的的說明圖； 疋根據本發明的—個顯示設計來處理吸收現象的方法的說明圖； 疋根據本發明的—個顯示設計來處理吸收現象的方法的說明圖； 穴圖33疋以用於本發明的—個顯示設計的全像顯示裝置為根據的顯示 貢料所採取的路徑的說明圖； 觀察者視窗錢嶋細錄虛擬 【實施方式】 A. 了在和像細在的同—基板上進行計算的全像影像顯示器 ^的-個顯示設計’包括有—個顯示器，可以接收真實空間影像 應資料於-個三維立體影像的—個光強度對蹄料和—個緩衝區對 近即時的韻全像編碼接著會根據三維立體影輸以即時或 單元/枯出來。透過結合兩個功能單元，也就是“全像影像計算 D全細_單元’’（這些單元與辦物裝置在功能上 23 200916986 ==都有所分別’以便在—個基板上構成—個共同的單元），全像影像 w、一^"至）有—部份可能在像素矩陣所在的物理空财執行。這表示至 二曰雜用於全像影料算的電晶體可以整合在用於進行像素控制的各個 Γ體Γ間或相傍並排。此外，透過結合兩個功能單71，也就是‘‘全像影 像十#早70 #全像影像顯示單元，，（這些單元與先前專利中的裝置在 功能上她山卩麵㈣，鄉—簡观—做轉元、 有的全像影料算也可能在像素轉所麵物理空財執行。或者， 份或所有_行全細___㈣_轉之外啊 和用於進行像素控_電晶體所麵同—基板上。義悉這類技術的人來 說’應該可以很清楚所謂的“在同—基板上，，並不表示這些電晶體只备在 原子層級塊嫩，術满纖墙了_支物 介’並在這崎繩相_路崎。“基板，，的意義的進—牛 資訊將在題名為“基板，，的—節中加明述。 " …在像素矩陣中（或者是在所稱的同_基板上）所進行的全像影像料 异，並不限贱祕細輪恤細晴枝。其他種類的 什异方法’像是“檢視表”（LcxMp Table，LUT)方法，也可以進行。 —個分析树输侧峨—域輸紐蝴方法。^ 素矩陣中的全像影像計算方面，整_示紅所有的全像運算方法可能都 完全相同，並且胁加人次全像影像崎料偏好在—個次全像影像尺 距離上進行交換。妓像影像料被料餅縣。料叫質^ 散在整顯示器表面上。但為了讓硬體的設計、模擬、和驗證更方便，^ 24 200916986 個 能將運算分誠鱗多小而完全相同的部份，稱為“叢集”，拼貼在顯示 器表面上。這些拼貼的圖點並不—定要是矩形的構造，也可以是其他的構 造，像是六角形（“蜂巢狀”）的構i這個所謂的“叢集，，會用來做為 -個運算單位’它涵蓋—部份或者整個全像影像運算的資料路徑。所以— 叢集可能就是可以用來根據一部份的原始真實空間資料來計算出顯示器 上的一__全像影«料的最小單位。這«集偏好在互減鄰的單 位之間交換資料，使來自互滅鄰的單位的次全像影像相互重疊的地方也 可以在SLM上正確的進行編碼。我們在圖24即以義方式對此加以說 明。採用叢集方法的-個優點是’在將㈣設計好之後，就可以輕易的透 過將所有完全相同的叢集拼貼在—起來來建構成—個全像顯示器。 理論上’要顯示具有極高影像品質的全像影像需要有極高的解析度（例 如，1M00 X 12，_像素），或採用一到幾⑽寬而不是只有幾刪寬的 虛擬觀察者視窗’或者兩者同時韻。要顯示的影助容，包含有一個光 強轉像資訊和三維緩衝區資訊（這可以稱為—個“z軸緩衝區”），典裂 k、有個取夕2, 000 xl，500像素的解析度。如圖i中所示，用來顯 示全像影像所需哺料傳輸速率遠高於用來顯示縣倾所需的資料傳輸 速率’例如，罐输輪者㈣48倍。㈣帽示，三維 :體影像資料以-個光強度對應資料和一個三維緩衝區對應資料的形式供 應。我們偏好针對每-個眼睛（也就是針對每—個虛擬觀察者視窗）構成 人組成對的緩衝區對應資料和光強度對應資料。這些對應資料每—個都包 己有—個2’_ x U 5〇◦像素的資料陣列。每—個對應中的每—個像素的 25 200916986 資料以三個_和—個z值（即四個值，Μ 8㈣來表示。—個 位元就是-個二進位數元。所以每個像素需要有32位元。視訊資料以沾 Hz的頻率供應，即每秒25格（25 fps)。圖圖所示，若使用兩個視域（右 眼和左眼）’職料傳輸速率為每秒Μ⑽。這個資料相來計算全像 影像’在簡單的情況下會-格-格計算，不過在更複雜的情況下可能會執 他㈣侧酬猶s㈣，像妨咖娜、人工處理 痕跡，或者像是減少所需的資料傳輸速率等等。全像影像運算會產生對摩 於一個1M00 x 12，咖像素的資料陣列的資料輸出（其 位元表示，而影格速率為150fps，採用25&的視訊頻=^ 域和三種色彩）。因此，如圖所示，全像影像的資料傳輸速率是每秒23〇 用圖1的内容代表同時顯示出三個原色（紅、綠、藍）的處理。這個 ^疋Μ針對—位使时的配置，但若是針對多位伽者的配置（且 有相應車父向的顯示影格迷率）也是可行的。對孰朵 - 於全像顯的資料傳輸速率方面，還可以舉出=其他的範例來說’關 必須細的是，__25Ηζ的影格物於活動的影像來說是 盯接又的最低逮率。為了達到更流暢的播放，應使用高於 格速率。雜速率越高，觀看者所看到的播放將會越流暢^ Ζ、办 一個全像影像只能針對—織定的絲顯示波 是需要針騎-崎這也正 色（即紅、綠、藍m々 ，沈疋針對母—個組成顏 电成來產生，而4 次。其他㈣彩可《這三種色彩 ㈣物驗-侧騎蝴蝴來進行。 26 200916986 如果王像影像是在同—基板（例如，在像素矩陣中）上的電路組件中 產生的取有原始影像資料需要傳輸到顯示器基板上。如果全像影像是 利用像素矩陣中的電路組件產生的，則光強度和緩衝區資訊會傳送到面板 中務後需要使岐㈣絲_蝴觸咖m卩本發明的—個 二蝴根_偏好顯示財，在計算全像影像中的—個像素的值時， 它只會考慮原始影像中的-個次區段的值。這麼做的-個理妓，在以本 《明的-個顯不設計為根據的偏好顯示器中，用來進行重建所使用的光並 =是在整個顯示器的各個部位妓完全—朗，耐在顯示器的各個次區 段之間存術-致，·:峨观_的各個懒區段。顯示 器的-個次區段與顯示器的另—個不同輕段之間可能不存在有不—致 的現象，或者只有非常有限的不—致的程度。偏好的顯示器中的每—個次 ^又可“來產生整個全像影像中的—個相應的次全像影像。因此，一個 次全像影像的尺恤定—個蝴議域（需要從這裡取得原始影像 的光強度和緩衝區值來進行次全像影像的計算）的最大延伸範圍。而這也 會決定必要_部線路（也就是所謂的“區域互連線路，，）的長度：情參 閱圖3。根據這個解決方案，由於用來產生全像影像所需的大量録^ 的全部或至少—部份會直接在顯示《上將《騎顯_部位進行吁 權⑽㈣鳩_) _姐麵轉 顯不負料或進行資料的中間儲存。這可減少傳送到顯示面板時所需的資料 的解析度，並可因此減少傳送到顯示面板時所需的資料傳輪逮率。如果將 這個範例套用在則t所示的情況，則可以達到減低約5Q% _料傳輸速 27 200916986 率的效果。因此，遍佈在整個面板上的橫列和縱列線路（即 互連線路”，請參_)的量也會因此而減少。它只跡、王區 資料更少的線路即足以進行原始影像資料的傳輸，而且傳輸頻=像影像 處。 了減^列和縱列驅_置的電功率消耗的好 減低資料傳輸頻率且有可、、志小# 手”有了減％列和縱列驅動裝置的功率 2。這是咖-個：梅姆順―（输—赠為零 消耗號電神，·當__高時，電神_也會升高。這此功率最 題4料熱的形式排散，而這會為高資料傳輪頻率的顯示器帶來高溫的問 二ΓΓ問題可能包括零組件會糊危險的高溫、因在高溫下降低庫 懈撕_壞、獅必她學祕電子峨 化、液晶材料由於暴露於極踹、、田庳 一 度下而導致的品質魏、以及由於溫度升 =致的轉體材料行為的改變如“熱載子生成，，。喊裝置使用電池 采’如果耗用更多的功率，則電池的電量會更快耗盡，這會減少每次 電池充電後可供裝置持續使用㈣間。 專利的解決方案中因縱列和橫列線路所需的每個像素的面積 的一大部分現在可簡在其_目的上。κ 2味了這兩個解決方案的 用原理。在以先前的專利為根據的解決方案中，以一個具有16，剛χ 2’_素_析度全___。為縮_和縱列線路， :㈣成4 _行拼貼’如㈣中的範例所示。每—個象限有8, _ 條縱列線路和6，_條橫列線路；總計需要有32，刪條縱列線路和 28 200916986 24, 000條橫列線路。針 τ耵位使用者而s，兩個視域（右眼和左眼）各有 三種組成色彩（即RΓ ^ 、B)各以25 fps的視訊頻率（輸入資料—光 強度及Z軸緩衝區〜砧旦《 人 的々格迷率）顯不，會產生每秒15〇個影像的顯 ^率乘以；^列數並加上每讎格之間的⑽空白傳輸時間，需 的縱歹J驅動頻率。而在以本發明的一個顯示設計為根據的解決 案的们侧巾’騎讀會根據—個2，咖X ι,_像素的實際影像 像素陣列來供應。如糊示器也分成4象限進行麻，每—個象限有75〇 條松列線路。將這個數字乘以每秒15Q個影像並加上每個框格之間的· 空白傳輸時間，則只需要135 kHz的縱列驅動頻率，如圖中所示。這個範 ]疋用於’、針肖位使用者的配置，但若是針對多位使用者的配置（具有 相應較高的顯示影格速率）也是可行的。 根據所使㈣面板和相_計算參數，在以本發_—麵示設計為 根據的解决方案（如圖2)巾，軸在橫列和縱列線路方面所節省的空間可 此是u不足道但在和以先則的專利為根據的解決方案（如圖2)相較時， 部可此大於細全像影像計算的電路组件所需的空間，所以，僅只節省下 來的空間巾的-部料足以應制於崎全像影料算㈣晶體所需空 間。在這輯況下，可以增加翻電極的面積翻此可以改善LQ)的透光 度。因為是在所節省下來的像素面射進行計算的，因此，它已經不再需 要-個和顯示器不在同-基板上、並且在任何已知的傳統裝置中都造成相 當可觀的困難度和成本花費的額外計算單元。另外的—個優點是，事實上 它可以大幅減少面板控·複雜度，因為用於進行面板控_資料傳輸速 29 200916986 率大致上和傳統的LCD相同。一個2,000 x 1，500像素的典型解析戶、 視訊影格速率25 fpS和兩個視域、每個像素為32 bit的資料傳輸速率 為4.8 Gbit/s，這大約與一個具備6〇 Hz影格速率和三種8以七色彩的 1，920 x 1，600像素的TFT面板相當。這個範例是用於只針對—位使用者 的配置，但若是針對多位使用者的配置（具有相應較高的顯示影格速率） 也疋可订的。14表示這樣的—個面板使用傳統的顯示技術就可以輕易的進 行控制，反觀以如圖i中所示的旗bits/s的典型資料傳輸速率傳送整 個全像影像’包括在計算單元躺示職子m、以及在顯示器電子 系統與顯示面板之間的傳送，都只有在採用不僅在製做設計上有其困難、 而且也將所費不貲的特殊處理手段時才有可能實現；這對熟悉這類技術的 人來說將會大大的讚賞。 果我們饭6χ在空間光調繼上進行—個全像影像的二維編碼，其中 原始真實㈣影像有2,咖X丨，_像素，並以25 fps的視訊影格速率 2，則大致上需要有大約1億個電晶體來進行全像計算，也就是每個真 曰]像素⑥要有大約34個電晶體。這是就開/醜率為的單 a私路、’且件而έ。由於一個以多晶妙製成的抓的開/關頻率可能只有 1音25 MHz，因此大約需要有6億9千萬個電晶體（而不是前面所說的 1億個電晶體）才能彌補它較低的開/關速率。假設-個全像影像的解析 又:Q X 12’刪像素’這絲示每個全像影像像素大約f要使用4 、電曰曰體由於計算值只有在—侧的影像被齡時才能被寫入到像素 Ί巾因此每個像素還需魏外的丨到2個電晶體。如果轉相 200916986 同的解析度，則一個顯示哭的尺± °。、、越大，像素間距也越大，因此在—個像 素周圍可以額外排列的電晶體數目也會越大。更詳細的電晶體總數的估算 將在電晶體總數估算_節中加以說明。 如果面板是經由橫列和縱列線路來進行控制，這些線路應隨著顯示器 尺寸的增大而加寬。這是因為線路材料的電阻係數是固定的，而，如果採 用固定的線路斷面積，職路的電阻將會隨著它縣度而成正比增大；由 於線路材料的電阻係數是固定的，如果採用固定的線路長度和厚度，則線 路的電阻齡與6的寬度献比。這表示在像素轉巾計算全像影像的方 法對於傳統的控制技術來說是有利的，尤其是在㈣以及高解析度的全像 顯示器方面。 —將TFT電晶體整合在一起，也就是將用於計算的電晶體與像素電晶體 —起佈置在同-個基板上，具有極大的好處。 乂麼做师卜的代價’將只有隨著電晶體數目的增加而可能帶來的較 大的故障機率。獨這可以透過咖—齡錯輯算方法來_，在這種 方法中，個別元件上的錯誤將只會對計算結果造成極小的偏差，而與完全 /又有任何元件缺失的情況下所獲得結果相差無幾。 計算將會在許多她鄰的運算裳置（稱為“叢集”）中進行，如圖2及 圖3所示。通f，運雛（叢集）的大小必須進行最佳化設計，因為它 們的大小越大，-方面在資料傳輪速率上所能產生的節約效果越小，但另 —方面卻也更容易進行計算。 在本發明之顯示設㈣—個進—步的範例中，—_示器會用來顯示 31 200916986 根據真實_騎_來像影_，如__料和緩衝 區對應聽。編專·_财的—個先天_就妓們需要使 用沒有與_«馳件佈置铜—基板上的電馳件。麵額外的電路 組件必須佈置麵㈤餘之相—個·板上。這轉致一政不利 的性質，像是較大賴置體積和重量，消#者永恆追求㈣是妙、薄、 紐、小的顯示裝置。以本發個顯示設計為根據的全像顯示器的運算 電路組件則是在與顯示電馳件所麵同—基板上。這些運算電路组件可 此在顯4的各轉叙間，或者㈣在齡㈣録_ 同一基板上。 在 關於石夕液晶（Lc〇s)顯示n中的整合問題的說明： 這種情況和佈置在—個單晶.片上的小型_顯示器有此不同。 ㈣本軸顯示技術將可椒高_率，取即使是每個 =觀㈣來酿賴。财爾_細分都蝴 ;斤=Γ ’ 裝置將只仙鱗雜餘种斷。因為料進行計算 而、1面積仍然-樣，因此所能產生的節約效果事實上是因為 輪或储存更少物4量所造_。這會減少用於橫列和 積並有利，趣.科，祕_ ^的面 在的m 柯，初t路組件所 、土反上（但運异電路組件並不設在顯示電路組件之中），因在、> 解決方案將會比運算電路組件位於和顯示電路組件所在的不同基 32 200916986 決方案更為精巧緊凑且更為經濟。 區域轉送 由於已經存在有-個用於計算資料區域轉送的額外邏輯存在，它同時 ^可:用轉送原始影像給侧_域，這使得全區制和縱顺路完全 變成多餘。例如，原始資料將會透過—轉移暫存器從叢集轉送到其他叢 集。由於橫列控制是在本區進行的，因此省略橫列線路使得顯示器的右側 和左側也可以用於寫入資訊。 容錯運算裝置 “般的TFT顯不器大概都有（例如）丨，_ X 1屬像素的解析 又 u衣^上的錯誤存在，最明顯的就是所謂的像素錯誤。用於 全軸示的高解析度顯示器會有更高的像素數，因此也會有更多的抓數 .笞^曰大中田提阿出現像素錯誤的可能性。如果整合有額外的TFT來進行 使一個故_奶的錯^Γ。秘得它雜料細序設計為可 能之間極小的局部偏差。、到整個社，而只會造成與理想性 觀看者肉眼無法察覺、或者人類視覺 種情況下人們可此可以任受這樣的缺 某些製造上的錯誤报可能會造成 系統只會勉強感覺_的後果。在這 33 200916986 失。但是，如果是—個（例如）完全損壞的叢集則是人們完全無法忍受的， 因在這種情況下會有—些SLM的像素格會受到波及。 備用電路組件（像是TFT)可能會被製造在像素矩陣的空間中，使這 些電路組件在發現某些麟裝置啟_電路組件有轉時，可以用來取代 』某些在裝置啟動時會關的電路組件。—個裝置可能會隨時進行自我測 *、象疋測D式某彳電路組件的開/關特性是否顯示出這個電路組件故 障。故障的電路組件可能會在記憶體中（像是非揮發性記憶體）被記錄無 法使用，而其他的電路組件則被記錄為正在使用中且發揮正常功能。在“物 ^ ^ (Physics and the Information Revolution) ( J. Birnbaum 與R.S. Will聰，Physics T〇day期刊，2咖年一月，卯.洲犯）一 2 併在本文件中做為參考）中’已經針對容錯的傳統電腦電路組件報 、 麵似的方法。另外’電路組件也可能設計成會使將造成—個像素 、、交a的故障的可祕大於將造成―個像素永久變亮的故障的可能性， 因為後者對_看者將纽歧大_激效果。 針對最佳化的容錯設計方面’可能會採脉電馳件中居於較重要的 地位的較大元件尺相電晶體（尤其是具有歓的横向尺寸者）以減少電 路組件中較為重要的部件的故障機率。一個進一步的手段是混合計算管線 使-個有問題的單元所獲得的結果會被分布在一個較大的表面積上。如果 人們知道可能會在計算值中加人—個敍約為麵或以上的全像影像像 ” ^麼就可以可鱗解這—點。如果這些值全都來自相同的管線，一旦 &個皆線發生錯誤’朗有全像影像的像素值都會是完全錯誤的值。如果 34 200916986 -個叢集包含有並行的管線’職集的内部結構可以㈣來加人的值將來 自所有並行的管線的方式鋪。舉例來說，如果這些縣自4個管線，如 果其中的-個管線發生錯誤，則只有25%的輸入值會是不正確的。在這個 情況下’所計算的全像影像像素值將會比1_的輸人值全都不正確I 況更為精確。 在某些情況下可能會採用一個“後續修補，，的策略。在這樣的情況 下’人們可以在顯示器的測試階段辨認出有問題的單元，接著並以物理方 式切斷相_料祕祕改電輪件；職的—财法可㈣來解決短 路的問題。靖連麟路和確歸“衫錄，，的像素轉（例如持續a 以極高光度閃耀的像素）只需要簡單的將它們切斷（醜）讓它變暗就可 以獲得適當的改善。 ‘在以本發明的顯示設計為根據的裝置方面，裝置可以根據後面所述的 “製造程序概述”、《錄相組合、或根據絲賴技術的人所知道 的其他製雜縣製造。有齡導體也爛來製造財發_顯示設計 為根據的裝置中的電路組件。 B.可在同—基板上進行計算、可進行有效率的空間光調編碼計算的全 像影像顯示器 用於進行三維内容轉換以表現可即時或近即時更新的大型電腦產生全 像影像⑽）重建的已知方法都只有在運細方面花費極大的心力才有 35 200916986 可能實現。在先前的專利申請中所述的“透過 生全像影像的方法，，（第___號文件）的^生電腦產 ν 1ηοη . β又件）的改良方法中，有iq卯 (pc ^_編_树询彡㈣咖彻個人電腦 的方式g預先#的次全像影像並在檢視表⑽）_助下以互動 分散m恤術_蝴_崎戦在特定的 如圖14中以空心圓圈所示。而這裡所將說明的本發明的 :示設輪謝迴賴制，各嶋㈣在重建的二 =的任何位置產生，如圖14中·，所示。圖 用先則的專利的方法所產生的各個物件點（空心圓圈）如何固定的八 配到特定的物件平社。而物件平_明定的轉定位在全像影像= =。相反的，根據本發明的-個顯示設計的分析式方法，各個物件點（實 心圓圈）則可以在任何位置上。 、 前面A項的顯示設計可以利用先前的專利的方法實施來進行空間光 調制器的編碼計算。此外，前面A項的顯示設計也可以利用—個可提供更 有效率的《光翻義碼計算的方絲實施。在第wq麵聰咖號 申請文件中說财-個更有效率的計算方法。下顺有效率的方法（本身 並不須要進行傅讀轉換或___計算，可財效率的實⑹ 就是本申請人對本發明的-個顯示設計。也可以說下列更有效率的方法並 不須要進行傅立葉轉換或菲涅爾轉換的計算。 一個可以提供更纽率的空間光調制器編碼計算的方法的範例如以下 所不。它是-種絲供-個全像顯示裝置（_產生“電腦產生視訊全像 36 200916986 〜像的刀析式方法（配合參考圖8及圖9來說明），包括有—個⑽光 調制手段⑽1)^其中將由物件所發出的波前將會在—個或多個虛擬觀 不者視_ (vow)中重建’且其中—個三維景象（3D_s)的每個單—物件點 (〇P)的重建只需要—個次全像影像（SH)做為要在SLM上進行編碼的整 個王像心像（ΗΣΜ)的子集’特徵是在一個3D景象（邪—s)離散化 (Discretization)成為多個物件點之後，這個方法包括下列步驟： 針對3D景象的每個可見的物件點 步驟L針縣-健伽（Gp)蚊次全像影像⑽）的位置。 例如’利用父叉線定理”，其中一個虛擬能見區域從全像影像平面 經由物件點投射到SLM转。只要具有足夠的精確度，次全像影像 可以大致形塑成-個矩形。然後對次全像影像指定一個區域座標系 統，以它的中心點為原點；x座標是橫座標而y座標是縱座標二欠 全像影像的尺寸是：“a，，為寬度的一半而“b”為高度的一半。 步驟B :針對全像影像平面（HE)細⑽每—個次全像影像（sh)決 定虛擬透鏡（L)的次全像影像： B1 :決定虛擬透鏡的焦距（f) 透鏡的焦距⑴是全像影像平面（HE)中要進行重建的物件點⑽ 與SLM的垂直距離。 37 200916986 B2 :透鏡的私像影像（SIL)的複雜值： 次全像影像複雜湘τ财程式來決定200916986 IX. Description of the invention: [Technical field to which the invention belongs] - Benshe County is associated with a high-resolution illuminating image of a high-resolution image dragon that has been decompressed by a decompressed vehicle. - on the woven board. The high resolution display device can display the same image data at the same time; the high resolution display device can also be a display capable of displaying the holographic display data. [Prior Art] The holographic image generation (CGH) is encoded in one or more spatial light modulators (10); the SLM may include a number of gamma pixels that can be electrically or optically controlled. These pixels will modulate the amplitude and/or phase of the light by encoding the holographic image of the age-of-age image. CGH can be calculated, for example, by ''lighting the light', by the reflection of the light reflected by the landscape, or by the interference between the reference light waves, or by Fwier or Fresnel ( Fresnei) conversion to calculate. An ideal Μ will be 嶋 (3) tender record value, (10) 卩, can be Qingqing - the amplitude of the input light wave and her. But ... a syllabary SLM will only control - a property, not Amplitude is the phase, and there may be other effects that may affect another side effect. There are different methods that can be used to modulate by amplitude or phase (10) light, for example, electrical addressing two liquid crystal SLM, optical addressing Liquid crystal SLM, magneto-optical (10), micro-reflection: industrial modulator, etc. The modulation of light can be continually continuous or include multiple m-type arrangements, individually addressable (binary, multi-order or continuous) In this paper, the so-called "encoding" means that each area of the object is used for 200916986 to write a hologram image for each control value so that it can reconstruct a scene through the SLM.The pure autostereoscopic display differs in that through the video holographic image, an observer can see the optical reconstruction of the optical wavefront of a two-dimensional scene. 31) The scene is reconstructed in an observer's eye and spatial light modulator (10)) Among the spaces that are unfolded. The SLM can also use the video king image to enter the code so that the observer can see the reconstructed 3D scene in front of the SLM, as well as on the SLM or other behind the SLM. The pixel of the spatial light modulator is preferably a permeable pixel that allows light to pass through, and its light can be interfered at at least a defined position and with a length (a few left-handed illusions). Allows the holographic image key to have a resolution of __# in at least _ dimensions. This ray we call it "sufficiently consistent light." In order to ensure sufficient consistency over time, it is emitted by the light source. The spectrum of light must be in a suitably narrow (four) wavelength range, that is, it must be close to the monochromatic I high-brightness illuminating - the spectral bandwidth of the light emitted by the polar body (LED) is narrow enough to ensure full-image Reconstruction in time. The diffracting limb at the SLM is proportional to the wavelength, which means that only a monochromatic source will produce a ''reconstruction' of the object. A wider spectrum will result in a coarser object point. And the object of « will be unclear. - The light of a laser is considered to be monochromatic light. The line width of the LED light is narrow enough in space to promote good reconstruction. Sex is related to the lateral width of the light source. Traditional light sources, such as coffee or a cathode camplight (10) L), if _- ____ line, can also meet these counties. - (d) light corrected by the source It can be seen as a light source emitted by a point source within the diffraction limit of 200916986, and, according to its "morphological purity", can be used to reconstruct the object of the paste. That is, each - An object transfer is keyed to a point within the diffraction limit. The light emitted by a source that is inconsistent in the work room will have a large lateral width and the reconstructed object will be blurred. This mode is due to the fact that it has been rebuilt at a specific location. In order to re-use the light source in the upper part of the space for the reconstruction of the full-length shirt image, it is necessary to find a -coin point between the brightness and the lateral width of the wire by a light-transmitting hole. The smaller the light source, the better its spatial uniformity. A linear light source can also be regarded as a point light source if it is viewed from the angle perpendicular to its longitudinal direction. In this direction, light waves can propagate consistently, but they are inconsistent in the other direction. On the whole, a holographic image will be transmitted in a holographic manner by a female image in the horizontal and vertical directions of the light wave. The photographic image (4) is a "full parallax holographic image". The objects it reconstructs can be viewed horizontally and vertically and have motion parallax, just like a real object. However, if you want to have a wide viewing angle, you must have a very high resolution in both the horizontal and vertical directions of μ. Throughout the system, the requirement for SLM will be limited to a “horizontal parallax only,” (全ρο) holographic image; it will only perform holographic reconstruction in the water, but not in the vertical direction. A holographic image reconstruction will be performed. This will result in the reconstructed object having only horizontal motion parallax and the vertical perspective of the reconstructed object will not change. A Ηρο hologram image has a resolution requirement for SLM in the direction of vertical 7 200916986. - A full-parallax holographic image is low. A holographic image of "Vertical Parallax Only" (VP0) is also possible, but not common; this holographic image reconstruction will only occur in the vertical direction, thus making the reconstructed object There is only vertical motion parallax, but there is no motion parallax in the horizontal direction. Therefore, different perspective landscapes seen by the left and right eyes must also be established separately. Instant calculation of holographic images requires very powerful computing performance, currently It can be implemented by expensive and specially manufactured hardware devices, such as component programmable logic (FpGA), π kingd 1C, or application-specific The device of circuit (10) G) is calculated using a plurality of central processing units (GPUs) that can be processed in parallel. On thin film transistor (TFT) displays, the pixel pitch in mutually perpendicular directions determines the area of each image 2. The area is divided into transparent electrodes for liquid crystal (10) control, _ M and column and traverse. The number of lines required for riding and hiding and the size of the display will determine the required shape, so it will also determine the horizontal The width of the column and the column line. The total transmission of 2 is higher than that of the current supply of the monitor device == 鲜 物 细 ^ 。 。 ( ( ( ( ( ( ( 比例 比例 比例 比例The available area of the large two-two transparent electrode is greatly reduced, and the transmittance of the display is significantly limited. The high resolution full-image display of the frequency ====^_ frequency is only in strict line holography = the performance requirement Extremely high, it can be used to access the real-time (10) axis, relying on the hardware device of the object-峨, 200916986, mussel. Because it involves a lot of data, the transmission of image data from the computing device to the display is also very Difficult. A general explanation of the common structure of an active matrix liquid crystal display device will be briefly described with reference to the prior patent documents (see _1G, US Patent No. 6,153,893); US Patent No. 6,153,893 is incorporated herein by reference. For reference purposes only, as shown in Figure 1〇, the active matrix is programmed (4)—the plate is made, the plate is like the plate (8), the plate is off the substrate, and the main substrate and the opposite substrate are fixed. The space 103 is formed, and the liquid crystal on the surface of the paper wheel (4) constitutes a display portion (10), and the portion of the pixel electrode iq4 and the wire electrode 104 of the wire array are opened/closed, and The peripheral-partial open/close device connected to the display portion 106 is composed of a thin film transistor; and the thin film transistor is also included in the peripheral portion 107 to constitute a circuit element. The W0 2 Gang/〇_6 file proposed by the Applicant (combined in this document) is a method for generating a holographic image by the computer. According to this method, the point matrix of the parallel virtual cross-section layer is divided into a parallel matrix of virtual cross-section layers, so that each of the cross-section layers has a meaning of a discontinuous amplitude value in the point matrix. The data set of the object is calculated for the holographic image of the _ holographic image __光_. . According to the applicant's paper No. 25839 (incorporated in this document for reference), the following steps will be carried out through the assistance of _: 'Parallel to each section layer at a limited distance The observer, according to each of the 200916986 fault scenes, calculates at least ~ _ 〃 in the form of the high cloth. 4 All the sections are calculated for the virtual observer window of the observer plane window located near the eyes of the observer. _ wave field, the aforementioned observation area is smaller than the video holographic image; the surface-related distribution of all the layers calculated by the computer is used to view the side of the face in the face of the mei-Gu fine;...' Converted into a holographic image at a finite distance and parallel to the reference plane: the rate ιΓ is generated for a scene-collecting computer to generate a holographic image--a full-image after the image is set in the full-length _ And after the encoding, the reconstruction is performed in the space in front of the observer's eyes through the assistance of the spatial light modulator. The method and display mentioned above are based on the idea of not reconstructing the scene object itself, but reconstructing the virtual observer window emitted by the object itself in one (four) = wavefront. The observer can fine-tune the age of the wheel. she was. The fun window is covered with the pupil of the observer's eyes and can be traced through the known position _ and the help of the money to trace the position of the sneakers... _, _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Run coffee and contempt (four), its t su ^ the bottom surface of the frustum, and the observer window represents its top surface. If the observer window is very small, this frustum may be close to a pyramid shape. The observer looks through the virtual observer window toward the display and receives the wavefront representing the scene at the observer. The amount of necessary conversion is very large, and the omni-encoding process can cause a huge computational load. Instant coding will also require 200916986 to fly expensive high-performance computing devices. The applicant's proposed w〇细8/〇2583 «There is a cushioning „3D shadow material instant production can make people have to use the method of cutting the calculus n. This makes this Shen Jiansheng _ reading the image It is possible. * The document of the W〇2〇_2_, which is released by the Hugh Moonman, reveals that a computer can produce a full-image martial arts method. Through the construction __# / The individual object points ==t _ xiao _ _ « 峨 _ information ~ like the feed into the code. Similar to the m method specified in the previous document / 025839, the first method of the W0 /... The same is not to reconstruct the scene object itself, but the virtual view that will be emitted by the object itself before; the first modulated wave field will pass through - the empty _ device μ from the fully consistent Xie The three-dimensional scene of the age is reconstructed through the interference of the virtual slii in the space. The virtual observer window will be generated in the reconstruction space of each frustum shape with ', , and 4. This discriminating window will fall close to the observer's position, and can be traced through the location of the known location. Invited to go to the office, the show, the death of the m is the method disclosed in the π==ΓΜ5δ39 document. The area where the bribe sees a scene is stretched from the SLM _______ cone Blocks, heads, and successes (four) = = Γ one method is also based on a single-object point reconstruction only needs to be a sub-butter image as a sub-batter. Therefore, with each scene _: #讯 will not be divided into 200916986 to the entire holographic image, but only the full image, in the middle. According to this concept, in the scene _ 疋 & in the domain, it is called the second 6 λ 1 斿 main ^ The individual object points in the sequel will only be transmitted through the limited pixel area on the SLM. The view revealed by the _ number file is a two-two omni-directional image. Each object point of the reconstruction of the first milk can be 全 全 影像 《 于 于 于 于 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全 全Example, (4) Nasaki. Each view_distribution (4) is in the position of the inspector at the position of the inspector. In the pre-viewer's eye, in the preparatory processing step, | a decomposition becomes a lot of visible object points. These data are also self-explanatory. The processing steps disclosed in the two-dimensional W〇2〇_2_ file include: Interfaces. The position and range of the first step image and its y coordinate and find the hologram image of each object point [the full-image image will be derived according to the position of the object point, that is, according to (4) horizontal/rushing area distance. - Step 2: Capture the configuration of the corresponding sub-holographic image from the view. Step 3: Add these sub-images to repeat the above two steps for all object points, and to reconstruct the entire scene to form an overall holographic image. 12 200916986 According to a simple example disclosed in Document No. W0 2_/025839, the size assigned to an object is determined by the size of the county. The viewer's window of the money boring or its hodge is projected through the object point to the holographic image (4), which is projected onto (10). Based on this, the pixel metric of the sub-holographic image required to reconstruct this scene point is determined. According to the point-of-step view disclosed in Document No. 25839 of WG 20, a singularity will be applied to each hologram or total holographic image. For example, the hearing compensation is due to its position. The resulting SLM error, or used to improve the quality of the reconstruction. For example, 'Add the correction value to the data value of the sub-holographic image and/or the total hologram image. In addition, since each hologram image is defined by the viewer's actual position of f, it can be targeted to lesser observers (for example, if the viewer views the display from a side angle at a great angle) ) Produce some special inspectors. As stated in document W0 2008/025839, the principle of using a view table can also be expanded according to preferences. For example, the color of the lin and the bright parameters: the _ can be stored in a separate _. The data values of the corrected hologram and/or the total hologram may also be modulated in accordance with the brightness and/or color values from the view. The color representation is based on the idea that each primary color can be extracted from an individual view. The touch table preferences according to the method disclosed in the document WO 2008/025839 are produced in accordance with the patent document No. WO 2006/066906 or WO 2006/_19 (filed by the present applicant, incorporated herein by reference). The table is then stored in a suitable data carrier and storage medium. The general idea of the 200916986 example disclosed in Document TO 2008/025839 is illustrated by only one observer. The sight of a scene (8) will be based on the position of the observer (9) and the view of the viewer. __ is assigned _-jian_chao_(10)), and the window of the sap - The wave field without overmodulation is controlled by a simple (10) heart-峨, 魄, and fine-grained image values. This method and the display derived by this method are based on the idea of a wavefront from the reconstructed object in one or more virtual observations (()) of the object that does not reconstruct the scene. In Fig. 26A, the object is viewed as a single object point (pp). The viewer (9) can see the scene (8) through the virtual observer window (Simplified). The virtual observer «(10)) will cover the _ (9) _ pupil, and can track the __ position with the help of known position and tracking. By controlling the holographic image value of the holographic image to control the null stat (10) and thereby causing the wave field (which will be emitted from the display screens), by interfering in the reconstruction space. To reconstruct a three-dimensional scene. As can be seen in Guanzhong, according to the general principle of this display design, a single object point (pp) of the scene (s) will only be reconstructed by a defined pixel area on the spatial light modulator ((10)). Holographic image (10). As seen in the figure, the root=one-to-single solution's size of a sub-holographic image (SH) will be determined according to the cross-line, and then the reconstructed object point (〇p) The required pixel metrics. The position and extent of the sub-image (SH) is derived from the position of an object point (pp) (i.e., its (4) X, y-axis coordinates) and its buffer distance (i.e., Z-axis distance). Then, the holographic image values needed to reconstruct the inverse point (pp) can be retrieved from the view table (coffee). The human hologram (SH) is modulated with a brightness and/or color value, and then accumulated 14 200916986 to the holographic image plane at the individual locations to form a so-called "overall hologram". The data contained in the above view will be generated in advance. These data preferences are generated using the method described in document WO 2006/066906 (as described in the previous patents noted above) and stored in suitable media and age media towels. Through the help of the position and nature of the object, the corresponding sub-hologram and sub-hologram images can be calculated in advance, and the color and brightness values and correction parameters can be generated accordingly. Figure 26B illustrates this principle in more detail and shows sub-holographic images (10), SH2) assigned to respective object points m, P2). It can be seen in Fig. 26β that these sub-images are constrained and constitute an overall holographic image, that is, a small and continuous subset of the entire spatial tone (slm). In addition to defining the position and extent of the sub-holographic image according to the intersection line theorem, as seen in Figure 26, there may be other functional relationships. 3. Exploring the previous patents, the WO 2__59 (10) coffee / _994) and the first leaking patents (collectively incorporated by reference in this document) by the present applicant, A device that is used to reconstruct a three-dimensional scene; the device includes a point source or a linear source, a lens for focusing light, and a "light modulator. Unlike traditional hologram displays, this SIJ will be rebuilt in the transfer mode in at least a "virtual observer window" (for a discussion of this term and related techniques, see Appendix I and II) - a 3D scene. Each virtual observer window will fall close to the observer's eye and will be constrained to a size that causes the virtual observer window to fall into a single-diffraction stage' to allow each eye to be in a Μ surface Stretching into the virtual observer window 15 200916986 Frustal cone hybrid reconstruction of the empty three-dimensional scene reconstruction. In order to make a holographic image reconstruction completely devoid of money, the imaginary observer sees (4) the size of the money must exceed the periodic interval of the reconstructed diffraction stage. However, it must also be at least large enough for the viewer to see the entire reconstructed 3D scene through this window. The other eye can be reconstructed through the same virtual observer window, or the second virtual observer can be specified as t (generated by a second source based on the correlation calculation). Here... a visible area (which is usually quite large) will be limited to the virtual viewer window. The known solution solves the large area created by the high resolution of a conventional SLM surface, and weave it down to the size of the virtual observer window. Due to the geometrical factor, this has the effect of minimizing the diffraction angle, making the resolution of the current generation of SLM sufficient to achieve a very high quality instant holographic image reconstruction through a reasonable consumption level computing device. / The first 4/G22_物|] (all merged in this document as a reference road - a mobile phone 4 that can generate three-dimensional images, the dimensional image is the use of the automatic rape of the road : The problem with the automatic read-reading method - the pre-reading 4 - is usually the inside of the face-changer, and the focus of the viewer's eyes usually has the focus on the eyes of the viewer. Point 谕仏 & & 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 Significantly reduced. It may occur, or may be [inventive content] The high resolution of the data shows that the present invention provides a solution to the high resolution image 200916986, and the display includes a myriad of pixels, which are located on a substrate. On, the circuit components are in the same as their respective; their towels, and the _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ < works to perform decompression calculations, and provides decompressed high-resolution image data for subsequent display by each pixel on the display. According to the upper frequency, the crane ship is executed by a circuit component disposed on the same substrate on which the lying pixels are located. According to the above series, the solution of the shots is distributed in a distributed manner over the entire substrate of the pixel substrate. _ Above, think, shoot, to soft _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ According to the individual pixels 1 of the H-side side. In the above concept, the circuit component that can be used to perform the decompression calculation is arranged on each of the stalks including the shank (four). According to the reading, I thought that 'the towel will pass through the __ will send the coffee maker = all the coffee and coffee every 嶋 梅 对 对 = 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩 缩Jing Shi-like display information. According to the above concept, the display 其中 can display the hologram display material. 17 200916986 According to the above concept, the space used to perform the compression calculation is on the same substrate as the pixels of the display. The pixel is in accordance with the above concept, wherein the space for performing the compression calculation is on a different substrate from the display. Alignment According to the above concept, the cluster for performing the decompression calculation receives data via the display and the column line. According to the above concept, each of the clusters used for decompression calculation receives data via a parallel bus. According to the above concept, the towel is finely connected to the fine-aged axis-connection data. According to the above concept, Bessie is a very high-resolution display according to the above concept, in which decompression will be performed by each cluster for 40 ms or less according to the above concept, wherein the holographic image calculation will be Executed after decompression. According to the above concept, the leaf calculation performed by the code used to determine the spatial light _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Determining the leaf calculation performed by the encoding of a spatial light modulator does not involve the calculation of the Fourier transfQrm face-neutral transformation. Snel According to the above concept, the leaf calculation 200916986 performed by the mosquito-space light modulation (4) coding is performed by using the (four) fiber________. According to the above, the tapping of the ray-determining is performed in each of the furnace fields that are dispersed in the display, so that the coding of the pixels of the corresponding dispersed area is calculated for each of the dispersed areas. According to the above concept, according to the above concept, The circuit component includes a thin film transistor. The effective location of at least some of the circuit components is determined by the polycrystalline stone according to the above concept, wherein at least a part of the circuit component has an effective location from the continuous grain stone According to the above concept, at least some of the effective components of the circuit components are composed of polycrystalline slabs. According to the above concept, at least part of the circuit components have effective locations from single crystal eve According to the above concept, at least some of the effective position of the electric wheel member of the towel is composed of a single crystal grain. According to the above concept, at least part of the effective component of the circuit component is composed of an organic semiconductor. According to the above concept, the substrate is made of single crystal germanium. According to the above concept, the display device is manufactured at the time of manufacture. Liquid crystal technology. According to the above concept, the substrate is made of glass. According to the above concept, the video frame rate is at least about 25 Hz. 19 200916986 According to the above concept, there is a real ^ I- 'fe, the image will be transmitted to According to the above idea, the real empty composition. The green vertical W should be based on the above idea 'shooting (10) Wg 即时 wheel real-time operation. ==, where the holographic calculation will be performed _ see table method to perform. On the ups and downs, the money _: owe _ county holographic calculation. According to the above concept, 'shooting _ in manufacturing _ technology. According to the above concept, where the display will be painted in the display 靡) technology, the mosquito Propose - Street _ said (four) think of the high-resolution age of the device to display spears, kick decompression _ high solution of the financial image, including the following steps: (a) compress high-resolution image data; (b) transmission through the grinding High-resolution image data; (c) receiving compressed high-resolution image data; (4) performing face-down on the touch-high-resolution series (4); and (e) displaying the decompressed high-resolution image data on the display on. The present invention can be illustrated by the following preferred solid blue: Figure 1 is an explanatory diagram for showing that the data transmission rate of the holographic image is much higher than that of the original real space data; Figure 2 is the SLM of the prior patent. a comparison of the structure and performance characteristics of a part of the SLM that can be plotted in the space of the pixel matrix; Figure 3 is an SLM that can perform holographic calculations in the space of the pixel matrix. A part of the construction diagram of 200916986; a construction diagram of the SLM that can perform decompression calculations in the space of the pixel matrix for the holographic data display; Figure 5 can be in the space of the pixel matrix FIG. 6 is an explanatory diagram showing various cases in the manufacturing process of the TFT for performing decompression calculation for the display of the conventional evil display material; FIG. 6 is an explanatory diagram showing each case in the manufacturing process of the TFT; FIG. 8 is an explanatory diagram showing a method of reconstructing a hologram image according to a display design of the present invention, and FIG. 9 is a view showing a display design for reconstructing a hologram image according to the present invention. Figure 10 is a perspective view of a general construction of a conventional silk matrix liquid crystal display device based on a prior patent; Figure 11 includes a hologram based on a display design of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 13 includes an explanatory view showing respective manufacturing steps of an active matrix substrate in FIG. 11; FIG. 13 includes an active matrix substrate in FIG. Description of each of the further manufacturing steps; an explanatory diagram of a holographic display showing each object point at each knife position and at any position; 21 200916986 FIG. 15 is possible to provide based on a display design of the present invention The whole picture is the Cai factory. . FIG. 16 is an explanatory diagram of a view table for the secondary image image SH in the hologram display based on one display design of the present invention; FIG. 17 is for BRIEF DESCRIPTION OF THE INVENTION A diagram showing an additional processing unit for holographic conversion and encoding in a holographic display based on design; FIG. 18 is a view showing a holographic display based on a display design of the present invention, if An illustration of the sub-holographic image 'the computational load will be smaller (since the number of pixels is smaller). Figure 19 shows the scene displayed in time "the temple, another scene displayed in time, and FIG. 20 is an explanatory diagram of a holographic display device based on the present invention with an addressable (four) transmission capability; FIG. 21 is not a conference A portion of a spreadsheet in which the number of transistors in the hologram display based on the display design of the present invention is calculated; FIG. 22 is the remainder of the spreadsheet shown in FIG. 21; She has just completed her succession A schematic diagram of the path taken by a display omni-directional display device of the present invention for displaying data according to the present invention; FIG. 25 is an operation of the items displayed by the item Calculate a construction diagram that may be in the space of the pixel matrix, which is not a traditional 2D display data, or a holographic display of a data of a part 22 200916986; Figure 26 is based on the previous patent FIG. 7 is an explanatory diagram showing a method of reconstructing a hologram image according to a display design of the present invention; FIG. 28 is a panel panel based on a display design of the present invention. Figure 29 is an explanatory view of the geometrical consideration of ''absorption'); Figure 3 is an explanatory view of the geometrical consideration of "absorption,"; — a display according to the present invention An explanatory diagram of a method designed to process an absorption phenomenon; an explanatory diagram of a method of processing an absorption phenomenon according to the present invention; a panoramic image display device for a display design of the present invention is according to An explanatory diagram showing the path taken by the tribute; the observer window 嶋 嶋 嶋 【 【 【 【 【 A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A 'Includes a display that can receive real-world imagery data in a three-dimensional image--a light intensity on the shoe and a buffer to the near-instant rhyme hologram encoding and then according to the three-dimensional shadow Or unit / withering. By combining two functional units, that is, "full-image image calculation D full-fine_unit" (these units and the device are functionally 23 200916986 == are different) so that On the substrate constitutes a common unit), the holographic image w, a ^" to) some - may be performed in the physical empty money where the pixel matrix is located. This means that the transistors used in the holographic image calculation can be integrated in the side of each of the scorpions used for pixel control or side by side. In addition, by combining two function sheets 71, that is, ''full-image 10' early 70 # full-image image display unit, (these units are functionally related to the device in the previous patent (4), township-Jane View - do the transfer element, and some of the full image calculations may also be performed in the physical emptying of the pixel. Or, all or all of the _ lines are fine ___ (four) _ turn and used for pixel control _ transistor On the same substrate - on the substrate. People who know this kind of technology should be able to clearly understand the so-called "on the same substrate, does not mean that these crystals are only prepared at the atomic level, and the walls are full. _ 支物介' and in this akisaki phase _ Lusaki. "Substrate, the meaning of the advance - cattle information will be described in the title of the "substrate,, - section. " ... in the pixel matrix ( Or the holographic image made on the same _substrate, and is not limited to the fine twilight. The other kinds of different methods are like 'LcxMp Table, LUT. The method can also be carried out. - An analysis tree is the side-by-field method. In terms of image calculation, all holographic methods of the whole _ red show may be identical, and the threat of the holographic image is exchanged at the distance of the photographic image. The image is covered by the image. The county is called the quality ^ scattered on the surface of the whole display. But in order to make the design, simulation, and verification of the hardware more convenient, ^ 24 200916986 can divide the operation into small and completely identical parts, called "cluster" ", collage on the surface of the display. The points of these collages are not - the structure of the rectangle, but also other structures, such as the hexagonal ("homed") structure of the so-called "cluster," , will be used as an arithmetic unit 'it covers the data path of part or the whole holographic image operation. So - the cluster may be used to calculate a part of the display based on a part of the original real space data. __Full image of the smallest unit of material. This set of preferences exchanges data between units that reduce each other, so that the sub-images of the units from each other can overlap each other correctly on the SLM. Line coding. We illustrate this in a way that is illustrated in Figure 24. One advantage of using the clustering method is that 'after designing (4), it is easy to build by arranging all the identical clusters together. It is a holographic display. In theory, it is necessary to display a holographic image with extremely high image quality, which requires extremely high resolution (for example, 1M00 X 12, _ pixels), or one to several (10) wide instead of only a few. The widened virtual observer window 'or both. The shadow helper to be displayed contains a light intensity image and three-dimensional buffer information (this can be called a "z-axis buffer"), Split k, there is a resolution of 2,000 xl, 500 pixels. As shown in Figure i, the feed transfer rate required to display the holographic image is much higher than the data transfer required to display the county tilt. The rate 'for example, the tanker (four) is 48 times. (4) Cap, three-dimensional: The body image data is supplied in the form of a light intensity corresponding data and a three-dimensional buffer corresponding data. We prefer to form buffer correspondence data and light intensity correspondence data for each pair of eyes (that is, for each virtual observer window). Each of these corresponding data has its own data array of 2'_ x U 5 〇◦ pixels. 25 200916986 data for each pixel of each correspondence is represented by three _ and - z values (ie, four values, Μ 8 (four). - One bit is - two binary digits. So each pixel 32 bits are required. The video data is supplied at a frequency of Hz, which is 25 frames per second (25 fps). If you use two fields of view (right eye and left eye), the transmission rate is Seconds (10). This data is used to calculate the hologram image 'in a simple case - grid-grid calculation, but in more complicated cases, it may be carried out (four) side compensation s (four), like 咖 娜, manual processing traces, Or like reducing the required data transfer rate, etc. The holographic image operation produces a data output for a data array of 1M00 x 12, coffee pixels (its bit representation, and the frame rate is 150fps, using 25& The video frequency = ^ field and three colors). Therefore, as shown in the figure, the data transmission rate of the holographic image is 23 每秒 per second. The content of Figure 1 represents three primary colors (red, green, blue). Processing. This ^疋Μ is for the bit-time configuration, but if It is also feasible to configure the multi-bit gamma (and the display frame rate of the corresponding car-father). For the data transmission rate of the 孰------the other example, It must be fined that the __25Ηζ frame is the lowest catch rate for the active image. For smoother playback, the higher rate should be used. The higher the rate, the viewer sees the playback. It will be smoother ^ Ζ, do a holographic image can only be targeted - the woven silk display wave is required to ride the needle - Saki is also true color (ie red, green, blue m 々, sinking for the mother - a composition of electricity It is produced four times. The other (four) color can be performed by these three colors (four) object inspection - side riding butterfly. 26 200916986 If the image of the image is in the same - substrate (for example, in the pixel matrix) The raw image data generated in the image needs to be transmitted to the display substrate. If the holographic image is generated by the circuit components in the pixel matrix, the light intensity and buffer information will be transmitted to the panel and the 岐(四)丝 needs to be made. Butterfly touches the coffee The two-rooted _ preference shows that when calculating the value of a pixel in a holographic image, it only considers the value of the sub-segment in the original image. The light used for reconstruction in the preferred display based on this "light-and-not-design" is used to be completely--in all parts of the display, and is resistant to each sub-section of the display. Surgery--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- The degree to which the preferred display in the display can be "to generate a corresponding sub-holographic image of the entire holographic image. Therefore, a sub-holographic image of the size of the image" The maximum extension of the sub-image image from which the light intensity and buffer values of the original image need to be taken. This will also determine the length of the necessary _ lines (also known as "area interconnections,"): see Figure 3. According to this solution, due to the large number of recordings required to produce a holographic image Or at least some of them will be directly displayed in the display "Appreciation of the ride (10) (four) 鸠 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The resolution of the data, and thus reduce the data transfer rate required for transmission to the display panel. If this example is applied to the case shown by t, it can be reduced by about 5Q% _ material transmission speed 27 200916986 rate Therefore, the amount of the horizontal and vertical lines (ie, interconnects) that are spread over the entire panel, as well as the amount of _), will be reduced. It is only enough to trace the data of the royal area, which is enough for the transmission of the original image data, and the transmission frequency = image. The power consumption of the reduced column and the column drive is reduced, and the data transmission frequency is reduced, and there is a power, and the number of the column is reduced by the % column and the power of the column driver. This is the coffee-a: Mei Mshun - (transfer - gift zero consumption number of electricity, when __ high, the electric god _ will also rise. This power is the most problem 4 material heat form, and this will be high data transmission frequency The problem that the display brings high temperature may include the high temperature of the components, the high temperature, the lowering of the library at high temperatures, the smashing, the lion’s knowledge, the electronic materialization, the liquid crystal material due to exposure to extremes, and the field. The quality of Wei, and the change in the behavior of the rotating material due to temperature rise, such as "hot carrier generation, shouting device using battery", if more power is consumed, the battery power will Faster depletion, which reduces the amount of time that can be used by the device after each battery charge. (4) A large part of the area of each pixel required for the tandem and horizontal lines in the patented solution can now be simplified. Objective. κ 2 tastes the principle of the two solutions. The former patent is based on a solution with an example of having 16, χ 2 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ As shown in the figure, there are 8, _ column and 6, _ horizontal lines in each quadrant; a total of 32, deleted column lines and 28 200916986 2, 000 horizontal lines are required. And s, the two fields of view (right eye and left eye) each have three component colors (ie, R Γ ^, B) each with a video frequency of 25 fps (input data - light intensity and Z-axis buffer ~ anvil" The 々格迷率) shows that the display rate of 15 images per second is multiplied by; the number of columns plus the (10) blank transmission time between each frame, the required median J drive frequency. In the solution based on a display design of the present invention, the side towel 'riding will be supplied according to the actual image pixel array of 2, X, _ pixels. If the paste is also divided into 4 quadrants for hemp , each of the quadrants has 75 rows of loose lines. Multiply this number by 15Q images per second and add the blank transmission time between each sash. Only a column drive frequency of 135 kHz is required, as shown in the figure. This is used for the configuration of the user of the pin, but if it is configured for multiple users (with a correspondingly higher display frame rate) It is also feasible. According to the (4) panel and phase calculation parameters, in the solution based on the design of the present invention (Fig. 2), the space saved by the shaft in the course and the column line However, this is not enough, but in comparison with the prior patent-based solution (Fig. 2), the portion can be larger than the space required for the circuit component of the fine holographic image calculation, so only the savings are saved. The space-part material is sufficient to make up the space required for the crystal image. In this case, the area of the flip electrode can be increased to improve the transmittance of LQ). Because it is calculated on the saved pixel surface, it no longer needs to be on the same substrate as the display, and it causes considerable difficulty and cost in any known conventional device. Additional computing unit. Another advantage is that it can significantly reduce the panel control and complexity, because it is used for panel control. The data transmission rate is roughly the same as that of the traditional LCD. A typical resolution of 2,000 x 1,500 pixels, a video frame rate of 25 fpS, and two fields of view, each with 32 bits of data transfer rate of 4.8 Gbit/s, which is approximately one with a 6 Hz frame rate and Three 8-color, 1,920 x 1,600-pixel TFT panels are comparable. This example is for configuration only for the user, but it can be customized for multiple users (with a correspondingly higher display frame rate). 14 indicates that such a panel can be easily controlled using conventional display technology, and in turn, the entire hologram image is transmitted at a typical data transmission rate of the flag bits/s as shown in FIG. The sub-m, and the transmission between the display electronic system and the display panel, are only possible when using special processing methods that are not only difficult to design, but also costly; this is familiar to this. People of the class will appreciate it greatly. If we have 6 χ in the spatial light modulation, we will carry out a two-dimensional code of the holographic image, in which the original real (four) image has 2, coffee X 丨, _ pixel, and the video frame rate of 2 fps is 2, which is roughly needed. There are about 100 million transistors for holographic calculations, that is, each true pixel has about 34 transistors. This is a single a private road with an open/ugly rate. Since a scratch-on/off frequency made with poly-crystals may be only 1 tone and 25 MHz, approximately 690 million transistors (rather than the 100 million transistors mentioned above) are needed to make up for it. Lower on/off rate. Assume that the analysis of a holographic image is: QX 12 'deleted pixels'. This shows that each holographic image pixel is about 4, and the electrical body can be written because the calculated value is only when the image on the side is aged. Into the pixel wipes, so each pixel needs to be outside the two transistors. If the same resolution is phased out in 200916986, then a ruler showing the crying ± °. The larger the pixel pitch is, the larger the number of transistors that can be additionally arranged around a pixel. A more detailed estimate of the total number of transistors will be described in the Membrane Total Estimation section. If the panel is controlled via a row and column line, these lines should be widened as the size of the display increases. This is because the resistivity of the line material is fixed, and if a fixed line area is used, the resistance of the line will increase proportionally with its county; since the resistivity of the line material is fixed, if With a fixed line length and thickness, the resistance age of the line is proportional to the width of 6. This means that the method of calculating a holographic image in a pixel swath is advantageous for conventional control techniques, especially in (4) and high resolution hologram displays. - Integrating the TFT transistors together, that is, placing the transistors for calculation on the same substrate as the pixel transistors, has great advantages. The cost of being a teacher's will only be a greater chance of failure as the number of transistors increases. This alone can be achieved through the coffee-age error calculation method. In this method, errors on individual components will only result in very small deviations from the calculation results, and in the case of complete/defective components. The results are almost the same. The calculation will be performed in many of her neighbor's computational settings (called "cluster"), as shown in Figures 2 and 3. Through f, the size of the chicks (cluster) must be optimized, because the larger their size, the smaller the savings in data transfer rate, but the easier it is. Calculation. In the example of the display device of the present invention, the -_ display will be used to display 31 200916986 according to the real_riding_image _, such as __ material and buffer corresponding to listening. Edited · _ _ - a congenital _ we need to use the _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Additional circuit components must be placed on the surface (five) of the remaining layers. This turns to the unfavorable nature of a political situation, such as the larger size and weight, the eternal pursuit of the consumer (four) is a wonderful, thin, new, small display device. The operational circuit component of the holographic display based on the present display design is on the same substrate as the display device. These arithmetic circuit components can be used in each of the rehashing of the display 4, or (iv) on the same substrate. Description of the integration problem in the display of n in the Lithium liquid crystal (Lc〇s): This case is different from the small-sized display arranged on a single crystal. (D) The display technology of this axis will be able to take the high _ rate, taking even if each = view (four) to brew. Treasury _ subdivision is butterfly; jin = Γ ’ device will only be broken with fairy scales. Since the material is calculated and the area is still the same, the savings that can be produced are actually caused by the wheel or the storage of less material. This will reduce the usefulness of the cross-column and the product, the interest of the section, the secret _ ^ face in the m Ke, the initial t-way component, the soil reverse (but the transport circuit components are not located in the display circuit components) The solution, > solution will be more compact and more economical than the different bases in which the operational circuit components are located and the display circuit components are located. Regional Transfer Since there is already an additional logic for calculating the data area transfer, it can also: transfer the original image to the side_domain, which makes the whole area and the vertical path completely redundant. For example, the original data will be transferred from the cluster to other clusters via the transfer scratchpad. Since the course control is performed in this area, omitting the course lines allows the right and left sides of the display to be used for writing information. Fault-tolerant computing devices "The general TFT display has probably (for example) 丨, _ X 1 genre of pixels and errors on the u ^ ^, the most obvious is the so-called pixel error. For the full axis display The resolution display will have a higher number of pixels, so there will be more captures. 笞^曰大中田提阿 has the possibility of pixel errors. If there is an additional TFT to integrate, it is a mistake. ^Γ. The secret design of the miscellaneous material is designed to minimize the local deviation between the two, and to the whole society, but only to the ideal viewers can not be perceived by the naked eye, or human vision can be allowed Such a lack of certain manufacturing error reports may cause the system to only feel the consequences of _. In this 33, 200916986 lost. However, if it is a (for example) completely damaged cluster, it is totally unbearable, because In this case, some SLM pixel cells will be affected. Spare circuit components (such as TFT) may be fabricated in the space of the pixel matrix, so that these circuit components are found in some of the device startup circuits. When the device is turned, it can be used to replace some of the circuit components that will be turned off when the device is turned on. - A device may self-test at any time, such as whether the on/off characteristics of a D-type circuit component are displayed. This circuit component is faulty. The faulty circuit components may be recorded in memory (such as non-volatile memory) and cannot be used, while other circuit components are recorded as being in use and functioning properly. (Physics and the Information Revolution) (J. Birnbaum and RS Will Cong, Physics T〇day, 2, January, 卯. 洲) 2 and in this document as a reference) Traditional computer circuit components report, a similar method. In addition, the 'circuit components may also be designed to cause a pixel, and the fault of the fault of a is greater than the probability that the pixel will be permanently brightened, because the latter will be a big difference. Amazing effect. In terms of optimized fault-tolerant design, it is likely to take advantage of larger component-scale transistors (especially those with lateral dimensions) that are more important in the pulse-reducing components to reduce the more important components of the circuit components. Failure probability. A further approach is to mix the computational pipeline so that the results obtained by a problematic unit are distributed over a large surface area. If people know that it is possible to add a person to a calculated value - a holographic image of the face or above, then you can scale this point. If these values are all from the same pipeline, once & The line has an error 'The pixel value of the holographic image will be completely wrong. If 34 200916986 - the cluster contains the parallel pipeline 'the internal structure of the job set can (4) to add the value will come from all parallel pipelines For example, if these counties are from 4 pipelines, if one of the pipelines fails, only 25% of the input values will be incorrect. In this case, the calculated holographic image pixel values It will be more accurate than the value of 1_ is not correct. In some cases, a "subsequent patching," strategy may be adopted. Under such circumstances, people can identify the problematic unit during the test phase of the display, and then physically cut off the phase-to-material secrets to change the wheel components; the job-finance method can (4) solve the short-circuit problem. Jing Lianlin Road and the “Picture Recordings,” (for example, pixels that continue to shine at very high luminosity) simply need to cut them off (ugly) to make it darker and get proper improvement. In terms of the device based on the display design of the present invention, the device can be manufactured according to the "manufacturing procedure overview", "recording combination", or other miscellaneous counties known to those skilled in the art. It is also bad to make a financial _ display circuit components in the design of the device. B. holographic image display that can be calculated on the same substrate and can perform efficient spatial light modulation calculation for 3D content conversion The known methods of reconstructing holographic images (10) for large-scale computers that can be updated immediately or near-instantly are only a matter of great effort in the operation. 35 200916986 may be implemented. As described in the previous patent application The method of holographic imagery, (the ___ file) ^ sheng computer production ν 1ηοη. β 件)) improved method, there are iq 卯 (pc ^ _ _ _ _ _ _ _ (4) G # manner previously sub-holographic image and view table ⑽) _ under interactive aid dispersion shirt Intraoperative _ m _ butterfly Kawasaki Zhan a hollow circles shown in FIG. 14 in particular. While the present invention will be described herein, it is shown that the 嶋(4) is generated at any position of the reconstructed two=, as shown in FIG. Figure 8 How to fix the individual object points (open circles) generated by the method of the patented method to a specific object. The object is flattened and positioned in the holographic image ==. In contrast, according to the analytical method of the display design of the present invention, individual object points (solid circles) can be at any position. The display design of the previous item A can be implemented by the method of the prior patent to perform the coding calculation of the spatial light modulator. In addition, the display design of item A above can also be implemented using a square wire that provides a more efficient calculation of the light ambiguity code. In the wq face Cong coffee number application file said that the financial - a more efficient calculation method. An efficient method (which does not require a read-to-read conversion or a ___ calculation itself, but a practical (6) is a display design of the present invention. It can also be said that the following more efficient methods are not It is necessary to perform the calculation of Fourier transform or Fresnel transform. A method for providing a more complex spatial light modulator coding calculation is as follows. It is a kind of silk supply-a full-image display device (_produced) The computer generates a video full image 36 200916986 ~ The image analysis method of the image (described with reference to Fig. 8 and Fig. 9) includes a (10) light modulation means (10) 1) ^ where the wavefront emitted by the object will be in Or multiple virtual views are not reconstructed in _ (vow) and each of the single-object points (〇P) of the three-dimensional scene (3D_s) needs only one sub-image (SH) as The subset of the entire image of the image (ΗΣΜ) to be encoded on the SLM is characterized by the following steps after a 3D scene (Dis- s) discretization becomes a plurality of object points: Every visible object point The position of the L-County-Giga (Gp) mosquito hologram (10). For example, 'Using the Parental Line Theorem,' one of the virtual visible regions is projected from the holographic image plane through the object point to the SLM. As long as there is sufficient accuracy, the sub-holographic image can be roughly shaped into a rectangle. Then specify an area coordinate system for the sub-images with its center point as the origin; the x coordinate is the abscissa and the y coordinate is the ordinate. The size of the under-image is: "a, half the width. “b” is half the height. Step B: Determine the sub-holographic image of the virtual lens (L) for the holographic image plane (HE) fine (10) per hologram (sh): B1: Determine the focal length of the virtual lens (f) The focal length of the lens (1) is the vertical distance of the object point (10) to be reconstructed in the hologram image plane (HE) from the SLM. 37 200916986 B2: Complex value of the lens's private image (SIL): Sub-image complex Xiang τ financial program to decide

ΐΑπ/八 I 八中人是光的參考波長，而f是焦距。等式中如果f有一個正 的值則表不相個凸面透鏡，域9A中所示。如 一個虛擬發散透鏡在SLM姆於觀看者的—側 = ⑽’如圖27中所示，則f«—個負 個物件點 β3:由於a在x和y方面的正負值是對稱的，因此只需要決定 一個象限中的a的值再利用適當的正負號將結果代人到其他： 個象限中即可。 、— 步驟 C.決定全像影像平面（HE)中的稜鏡的次全像影像⑽）： 由於所選擇的區域座標系統，加人—個稜鏡將會導致相位偏移，所 以相位偏移是x和y座標的一個線型函數。 ci:蚊具有水平效果的稜鏡⑻的線觀數Gx，關隔為h 〔〇 ’ a〕以内表示為ΐΑπ/八 I Eight people are the reference wavelength of light, and f is the focal length. If f has a positive value in the equation, then there is no convex lens, as shown in domain 9A. If a virtual diverging lens is at the side of the SLM - the side = (10) ' as shown in Figure 27, then f « - a negative object point β3: since a positive and negative values of a in terms of x and y are symmetrical, It is only necessary to determine the value of a in one quadrant and then use the appropriate sign to pass the result to the other: in the quadrant. , - Step C. Determine the sub-holographic image of the 全 in the holographic image plane (HE) (10): Due to the selected area coordinate system, the addition of a 稜鏡 will result in phase offset, so the phase shift Is a linear function of the x and y coordinates. Ci: The line view Gx of the 稜鏡(8) with a horizontal effect of the mosquito, and the interval between h (〇 ’ a) is expressed as

Cx = Μ * (2π/λ);其中μ為絕對稜鏡斜率（圖9B) C2 :決定具有垂直效果的稜鏡（p)的線型因數G，以間隔為r 〔0，b〕以内表示為Cx = Μ * (2π/λ); where μ is the absolute 稜鏡 slope (Fig. 9B) C2: determines the linear factor G of 稜鏡(p) with a vertical effect, expressed as an interval of r [0, b]

Cy = N * (2π/λ);其中N為絕對稜鏡斜率（圖9C) C3 .稜鏡的次全像影像（shp)的複雜值： 38 200916986 這個次全像影像（shp)的複雜值會根據棱鏡的重疊來決定，即 zp = exp { i*〔 Cx* (x-a) + Cy* (y-b)〕} C4 :如果光源是透過全像顯示裝置投射到v〇W上，則稜鏡修正可 以忽略。 步驟D:調制透鏡及稜鏡的次全像影像： 所結合的次全像影像的複雜值會透過—個虛擬透鏡（L)和虛擬稜鏡（p) 的效果的複乘法得出，如圖9A所示，如下 兄 zsh = zl木a，這也可以用SH二SHl木SHp來表示 步驟E:相位偏移 每-個次全像影像（SH)會以-個（統—分布的）相位偏移進行調制（這 裡的相位偏移每個次全像影像都不儘相同），以便能夠在能見區域中達= 均質的光度。這可以減何能因絲在光學上的__致性所造成的光斑。 相位偏移的幅度即足以減少細£，並且可能小於π弧度（亦即，並不一 疋要疋π〈 φ。&lt; π，而可以是（例如）_π/4〈 φ〇〈 π/4)。這個過 可以由下列等式表示： zsh := Zs„ exp (ιφ〇，這也可以用 SH SH exp 來表示 步驟F:光強度調制 各個次全像影像個觸複雜值會以—個根據影格緩衝區内容（單色，或 39 200916986 呈現各自她 蝴，細物件點可以 ZSH — C ^ 城這也可以用SH := c * SH來表示； 步驟G加人_人全像影像來構成整個全像影像 ΗΣ SLM ° 各個次全像影像可叫職加 次全像影像的複雜總和，以下列#^=整個全像影像就是各個 這也可以根據整個全像影像的-個座標系統用 、上述的轉,C、D、和E在以本發明的顯示設計為根據的某些範例中 (在允騎低運算的能力或全像影像的品質的情況 硬體製叫卿爛低為嫩蝴算方法所需的 ==意的是’如果重建的物伽被當作是—麵學系統的焦點， 这表不在王像影像平种有—個透鏡，而前述的這個透鏡有—個傾斜角度 而且它的焦岭f。—個傾_繼—個糊物辦-個棱鏡所 、=。根據喊所麵的方法…條件騎以可在—個次全像影像的編 碼中加入-個透鏡函數和（必要時）—個稜鏡函數的方式進行重建（請參 關9A)。-個由無數個點所構·景象可以透過疊加多個次全像影^產 生。透過朝這個方法，—個互動式即時全像影像重建的各個物件點可以 200916986 利用市售的標準硬體零組件在重建的截獅體空•的任何位置產生。过 個解決方銷時在鱗點的數量方面也可以輕“麵娜，只要提 理早疋的性能’就可以增加物件點的數量。 门 叶鼻的程序可以歸納如下·· 1.透鏡的計算 a. 找出焦距f b·使用透鏡等式：e]力㈤⑴仏 2.進行稜嫩，域卿辦） a.決定 &amp;、Cy、a、及 b b·等式：e、{i*〔 r CX (x'a) + Cy* (y—b)〕}Cy = N * (2π/λ); where N is the absolute 稜鏡 slope (Fig. 9C) C3. Complex value of the sub-hologram (shp) of 稜鏡: 38 200916986 Complex value of this sub-image (shp) It will be determined according to the overlap of the prisms, ie zp = exp { i*[ Cx* (xa) + Cy* (yb)]} C4 : If the light source is projected onto v〇W through the holographic display device, then 稜鏡 correct can be omitted. Step D: Modulating the sub-image of the lens and the hologram: The complex value of the combined sub-image is obtained by the complex multiplication of the effects of a virtual lens (L) and a virtual 稜鏡 (p), as shown in the figure. As shown in Fig. 9A, the following brother zsh = zl wood a, which can also be represented by SH two SHl wood SHp. Step E: Phase shift Each time the hologram image (SH) will have a phase (unit-distributed) phase The offset is modulated (the phase offset here is different for each hologram) so that it can reach a uniform luminosity in the visible area. This can reduce the amount of light caused by the optical __. The magnitude of the phase offset is sufficient to reduce the fineness, and may be less than π radians (ie, not necessarily 疋π<φ. &lt; π, but may be (for example) _π/4< φ〇 < π/4) . This can be expressed by the following equation: zsh := Zs„ exp (ιφ〇, this can also be used SH SH exp to indicate step F: light intensity modulation each hologram image touch complex value will be based on the video buffer District content (monochrome, or 39 200916986 presents each of her butterflies, fine objects can be ZSH - C ^ City This can also be represented by SH := c * SH; Step G plus people _ human hologram to form the whole hologram Image ΗΣ SLM ° Each holographic image can be called the complex sum of the holographic image, with the following #^=The whole hologram image is the same as the coordinate system of the whole hologram image. , C, D, and E are in some examples based on the display design of the present invention (in the case of the ability to ride low-calculation or the quality of a holographic image, the hard system is called the low-cost method) == means that 'if the reconstructed object is treated as the focus of the facet system, this table is not a lens of the king image, and the aforementioned lens has a tilt angle and its focus Ling f.—a tilting_following-a paste-doing a prism, =. According to the method of shouting... conditional riding can be reconstructed by adding a lens function and (if necessary) to a 稜鏡 function in the encoding of the holographic image (see 9A). Scenes constructed by a myriad of points can be generated by superimposing multiple sub-images. Through this method, an interactive real-time holographic image reconstruction of individual object points can be used in 200916986 using commercially available standard hardware components. It is produced at any position of the reconstructed lion body. When the square pin is solved, the number of scales can also be lightly “faced, as long as the performance of the early squatting” can increase the number of objects. The procedure of the nose can be summarized as follows: 1. Calculation of the lens a. Find the focal length fb. Use the lens equation: e] force (five) (1) 仏 2. carry out the tenderness, domain Qing) a. decide &amp;, Cy, a, And bb·equation: e, {i*[ r CX (x'a) + Cy* (y-b)]}

Cx = ( 2ττ/λ ) * mCx = ( 2ττ/λ ) * m

Cy = ( 2π/χ ) &gt;k n 3·進行棱鏡和透鏡方面的調制 I不一疋頊要，依處理過程而定 .套用任思的相位（不一定須i + 疋肩要，依處理過程而定） 5.光強度調制 6·進行全像影像的SLM指定編碼 5十算的全像影像顯示器 c.可在同一基板上進行解壓縮 41 200916986 於^發明的―個顯示設計包括—個可以接«實帥影«料（如對庫 於一個三維立體影像的一個光 子應 干考f a 強度對應貝科和—個緩衝區對應資料）的顯 ”器接者，會根據三維立體影像資料以即時或近 光調制器的全像影像_。@ 、、；计鼻出空間 干叶曾單元，，彳相私_魏單元（纽是“全像影像顯 示單元，，），所有的全像影像顯示器购至 崩的-物以槪__輪娜進 先二 專利中的裝置在功能上和空間上都H“ ㈣以敕^ w王像讀顯不計域至少其中-部份的電晶 I ^在祕進行像素控_各_«之岐树麟。此外，全 =計=，位於與像素電路組件所在的同一基板上的電路 組件中。丁 “二用於全像影像顯示計算的電路組件不包含在像素電路 =發明的一個顯示設計的這個進一步的範例中，全像影像計算會在 象=陣所佔難__—個峨行。_伽具有可區域 存取的檢視表⑽）的優點（如第WG瞻咖9號文件所述），這可 Γ各撕算的運算效率。如圖1所清楚顯示的，在顯示器像素空間以外 、乘行全像影像計算的手段的—個問題是需要有極高的麵料傳輸 :…傳送資訊給顯示器的各侧象素。但如果採用一個類似於如圖4中所 不的方法的+段，則可明免_敎。 在‘具不為中，全像影像的編碼資料會在像素矩陣所佔用的空間以外的 方進仃a十异。執行這些計算的空間可能是或可能不是在與顯示器基板所 42 200916986 的同基板上王像影像編碼資料會利用已知的資料壓縮技術進行壓 縮，之後再傳輪到顯示器的各個叢集（這些叢集是整個顯示器的—部份）。 在圖4中，用於進行全像影像計算的各_ τπ會對經由橫列和縱列線路所 接收到的資料執行解難的功能。*過，這些資料也可能經由其他手段接 收到’例如經由—個並列資料匯流排、或一個序列資料連接、線路。以一個 叢集-個叢集為基礎的全像影像齡$可以減少對各個全像齡器像素之 間的互連線路的需求，並因此可減少對影像光強度對應_和影像緩衝區 對應資料的來源的絲。全像影像計算柯料壓縮也可以麵示器基板以 外的位置上執行’其中龍解壓縮會糊位於顯示器的像素所在的同一基 板上的電路組件執行，但解壓縮則在像素矩陣的如以外的位置執行。對 熟悉這類技術的人來說，其他的範例自不必多加贊述。 D·可在同一基板進行解壓縮計算的高解析度顯示器 在本發明的一個顯示設計的一個進_步的範例中，一個高解析度顯示 器會用來顯㈣騎度影髓料’這可能是—錢示⑽或者也可能是根 據光強度對應資料和緩衝區對應龍所計算出來的全像影_示賴。以 先前的專利細麵高解析度齡器_有的問題是，它們都^極高密 度的電路組件，而這會更容易產生製造上 , _⑽切要極高的 開/關頻率，而這也可能會導致過度_ m心 Μ_似於圖5 所不的手&amp;的-個方法，則可以減少或避 43 200916986 在高解析度顯示器中，影像眘姐 像貝枓會在顯示㈣部或外部_已知的資 料壓縮技術進行壓縮，之後再傳&amp; 、 交丹得輪到顯不器的各個叢集（這些叢集是整個 顯不器的一部份）。進行壓縮計算的处 卫間了此疋或可能不是在與顯示器基板 所在的同一基板上。在圖5中，用於進行解壓縮計算的各個m會對經由 橫列和縱列線路所接收到的資料執行解壓縮的功能。不過，這些資料也可 能經由其齡段接_，勤經由—個並顺·流排、或—個序列資料 連接線路。為了減少對記憶體的需求，以一個25Hz的影格速率為例，用 於進行解壓縮計算的各個TFT將會被要求在約4〇贴或以下的時間内對這 個資料進行《縮來透過叢集的各個像素進行麻。以—個叢集_個叢华 為基礎的f彡像顯示料以減少對各個影像顯示雜素之_互連線路的^ 求，並因此可減少娜像光賊對應簡的來_需求。職悉這類技術 的人來說，其他的範例自不必多加贅述。 在-個較佳的範财，解壓縮後的真實㈣影像#料會傳送給顯示器 的各個叢集。在第—個步驟巾’叢齡執行雜·真實”影像資料的 解壓縮。销二個步驟巾，全像齡龍會由齡㈣各個叢集利用由第 一個步驟所產生的資料計算得出。對熟悉這類技術的人來說，其他的範例 自不必多加贅述。 E.可透過合併用於進行全像轉換和編碼的額外處理單元以一個擴充的圖 $子系統的邪著色管線(Rendering Pipeline)在同一基板上進行計算的 全像影像顯示器 44 200916986 刚面A項的顯示設計可以利用先前的專利的方法實施來進行空間光 °的扁馬此外，如面A項的顯示設計也可以利用一個可提供更有效 率的^間光棚病侧方法來實施。以下將提供—個可提供更有效率的 工間光樹!]&amp;編碼的方法的細，但對於許多熟悉這織術的人來說，其 他的範例自不必多加贅述。 ^ f 法（如圖15中所示的一個範例）’透過合併用於進行全像轉換 和編碼的額外處理單元，擴充了圖形子系統的3D著色管線(Rendering p ine)⑽方法就是本巾請人對本發明的—個顯示料。這裡所謂的 用於進仃全像轉換和編碼簡外處理單元”以下將以“完全管線，，這個 名稱來取代。補完全管線直接安排在3D圖形管線的下游。每個叢集所 ;.、&amp;、泉(PipeHne)資料會傳送給顯示器上相應的叢集；從現在起以 下的敘述將著重在—個單—叢集層級的顯示設計上。—個z轴對應緩衝區 和，色彩對應緩衝區（色舞應R、色賴應G、色雜應^構柄 個官線之間的銜接介面。請參照圖15中的簡圖所示。在像素座標中的每— 個個別的點方面’ z轴對應包含有—個z值，這個值有尺度並且可以以各 種，月晰度水準表示。2值的典型尺度是在介於〇 Μ 1.G之間的範圍 内’但也可能是其他的範®。它的清晰度水準會根據位元數麵定，即， 通常是8、16、或24 bit。 在現代的圖形子系統中，色彩對應的清晰度為24 bit，也就是每個分 P、’工’’、彔、藍）各8 bit。色彩對應構成影格緩衝區的—部 45 200916986 =、緩衝&amp;的内谷正常會顯示在晝面上。這兩個緩衝區（包含有z抽對應 =料和色彩對應資料）會加以定義來構成⑽著色管線細加啊 __和完全錄之_銜接介面。z⑽應資料會針對—個顯示波長 來提供，但R、G、B則沒有特定的波長。z抽對應資料的拷貝腿和· 則會分別針對另外__示波長來提供。 '一個全像影像只能針對—個特定_示献絲妨計算。這也正是 為什麼必彡_每—個物件點_三次計算_，也就是針對每個原色 錢订-人’紅⑽）、綠aG)、和藍（λΒ)。其他的色彩可以利用這三個 刀色創以出來’而這個色彩的調配可以透過循序或同步方式進行。為了提 间處理的速度’它會採用至少兩個額外的完全管線，使全像影像計算可以 亚订執仃’讓它可以同時得到所有三個分色的結果。為此，它必須將Ζ轴 對應貝料讀到額外的記憶體區段刪和簡（請參關⑸，這些區 •k可以獨立存取。因此，這可以防止涉及到記碰區段的操作（如ζ軸對 應貝料）可此會彼此造成妨礙的情況。所以，各個記憶體區段在理論上應 貫際分隔。色彩對應資料Μ中關於色彩G和B _容也會分別複製到 色祕應G、和色彩對應Β的單獨記憶體區段，以確保三個分色可以獨立 存取（請參_ 15)。同樣的，各個記憶體區段也可能實際分隔以防止在記 It體存取期間發生衝突並減少或甚至丨肖除雜的顯示設計巾帶有號諸 (maphore)、互斥運算（也就是“肌^exes”）等的同步存取問題，這些 問題可能對系統性能有不利的影響。然而，雖然記憶體區段可能彼此實際 分隔’但它們最好仍應位在顯示器的同—個叢集範圍中。請注意，一個號 46 200916986 諸是-個受賴的變數（或減職觸型）並構成在—辨元程式操作 環境中用來限制存取共享資源（例如，儲存空間）的典型方法；互斥運算 會使用在並行處理的程式設計中用來透過以電腦碼的片段叫出關鍵區段的 方式避免同時使用-個共享的資源，例如—個整體變數。 以下將假設-個全像影像由多個次全像影像所構成。其中的第以個次 全像影像^-個透鏡代表，這個透鏡可由—個透鏡函數說明Y㈠以 U⑽）。其中的常數G包括透鏡的焦距⑴的值會在套用透鏡函 數之前先行計算得出，使之後f的值可以用在所有的三個管線中。f的值 因此與色卿：峨㈠貞峨，目喻彡歸脱差。這可 以擁有透鏡函數_的優點，因為—猶鏡在它的χ麵y軸上是對稱 的。為了坪細說明一個透鏡，這個函數只需要套用在—個象限上。之後可 以將在個象限中所計算的透鏡函數的值利用正負號的對稱原則套用到盆 他三個象限上。 '' &amp;也會因波長Λ而異，三個原色（R、G、B)的波長本質上就各不相 H #值並不需要計算，因為，由於事實上使用的是—個絲的雷射或 、〜原來產生母-個波長，因此波長紅知的；不過，應在計算中代入又的 值以便計算顯示器的每個原色的Ct (請參閱圖15)。 根據所採物彳，珊了編#咖之崎瞻用一個棱 鏡函數（請參_ 15)以便改變光傳制方向。在棱鏡函射，一個常數 也包括有波長λ。這個f數的個此也會有所.，因為三_色各有不 同的波長，所叹财數的值在三歡全管線上每個管齡有—個指定的 47 200916986 值。 現在’透鏡函數和稜鏡函數在如圖15進行複乘法15〇3、15〇4、和 1505。接著，在套用隨機的相位1506、1507、和1508，加入到透鏡和稜 鏡函數的複乘法的結果中。這個方法的目標是避免在觀察者平面產生亮度 峰值或所明的光斑。接著使職別的色彩對應龍的光強度調制出 個別的全像影像1509、1510、和1511。 在下個步驟中’這個次全像影像會進行一項複加法來針對叢集構成 總體全像影像（請參關15)。現在，所獲得個結果可以利用在全像顯示叢 集中進行的加法運算來進行後續喊理（如果合適），例如，套用修正對應 或灰階影像（灰_正），由於這只會取決於Μ的綠性質，所以最好 在以_段進行修正。接著是編碼處理程序。全像影像可雜色彩來進行 重建。編碼運算法則（請參閱圖15)會因為所使用的Μ而有極大的差別， k可此疋依她編碼、絲幅編碼、或者依另—種方式進行編碼。 熟悉這類技術的人將會看得出來本節中所述的顯示設計中的某些方面 在本申請文件中的其他地方有更詳細的揭載。Cy = ( 2π / χ ) &gt; kn 3 · Modulation of prisms and lenses I is not a problem, depending on the process. Apply the phase of Ren Si (not necessarily i + shoulders, depending on the process 5.) Light intensity modulation 6 · SLM designation for holographic image encoding 5 omni-directional image display c. Can be decompressed on the same substrate 41 200916986 Invented the "display design including - can be connected «The real handsome shadow material" (such as a photon in a three-dimensional image of a photon should be tested to the fa intensity corresponding to the Becco and a buffer corresponding to the data), the device will be based on the 3D stereoscopic image data in real time or The omni-directional image of the low-beam modulator _.@,,; the nose-out space, the dry leaf zen unit, the 彳 phase private _Wei unit (new is the "holographic image display unit,"), all holographic image display purchase To the collapse of the object - 槪 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Perform pixel control _ each _« 岐 麟 。. In addition, all = count =, located in pixels The circuit components on the same substrate where the circuit components are located. D. "Two circuit components for holographic image display calculation are not included in the pixel circuit = a further example of the invention's display design. The hologram image calculation will be in the image. = Array is difficult __ - a limp. The advantages of _ gamma with a viewport (10) that can be accessed by a region (as described in the WG Vision No. 9 document), which can be used to calculate the efficiency of each calculation. As clearly shown in Figure 1, the problem with the means of multiplying holographic image calculations outside of the display pixel space is the need for extremely high fabric transmissions: ... transmitting information to each side of the display. However, if a + segment similar to the method shown in Fig. 4 is used, it can be clarified. In the case of ‘without, the encoded image of the holographic image will be different from the space occupied by the pixel matrix. The space for performing these calculations may or may not be on the same substrate as the display substrate 42 200916986. The image-encoded data will be compressed using known data compression techniques, and then transmitted to the various clusters of the display (these clusters are Part of the entire display. In Fig. 4, each _τπ used for holographic image calculation performs a function of solving the data received via the traverse and column lines. *Over, these materials may also be received by other means, such as via a parallel data bus, or a sequence data connection, line. The holographic image age based on a cluster-cluster can reduce the need for interconnects between the pixels of the full-aged pixels, and thus reduce the source of image light intensity correspondence and image buffer correspondence. Silk. The holographic image compression calculation can also be performed at a position other than the surface of the display substrate, where the decompression is performed on the same substrate as the pixel on which the display is located, but the decompression is outside the pixel matrix. Position execution. For those familiar with this type of technology, other examples don't have to be praised. D. High-resolution display that can be decompressed on the same substrate. In an example of a display design of the present invention, a high-resolution display would be used to display (four) riding shadows. This may be - Money Show (10) may also be based on the light intensity corresponding data and the buffer corresponding to the dragon calculated by the full image _ _ _. With the previous patent fine-faced high resolution ages _ there is a problem that they are extremely high density circuit components, and this will be more prone to manufacturing, _ (10) cut extremely high on/off frequency, and this may It will lead to excessive _m Μ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Known data compression technology compresses, and then transmits &amp; and dandan turn to the various clusters of the display (these clusters are part of the entire display). The server performing the compression calculation may or may not be on the same substrate as the display substrate. In Fig. 5, each m for performing decompression calculation performs a function of decompressing data received via the course and column lines. However, these materials may also be connected via their age groups, and they may be connected via a serial, or a sequence of data. In order to reduce the need for memory, taking a 25 Hz frame rate as an example, each TFT used for decompression calculations will be required to "retract through the cluster" in about 4 〇 or less. Each pixel is hemp. The f-images based on a cluster of clusters are used to reduce the number of interconnected lines that display the various images, and thus reduce the need for the image thief. For those who are familiar with this type of technology, other examples need not be repeated. In a better model, the decompressed real (four) image # material will be transmitted to the various clusters of the display. In the first step, the 'cluster age implementation of the real-life image data is decompressed. The two-step towel is used, and the full-aged age dragon is calculated from the age (four) clusters using the data generated by the first step. For those familiar with this type of technology, other examples need not be described in detail. E. Adding an additional processing unit for hologram conversion and encoding to an extended graph $ subsystem of the Rendering Pipeline A holographic image display 44 calculated on the same substrate. 200916986 The display design of the front face A can be implemented by the method of the prior patent to carry out the spatial light. In addition, the display design of the face A can also utilize a A more efficient method of illuminating the side of the shed can be provided. The following will provide a more efficient method of coding the lighthouse!] &amp; encoding, but for many people who are familiar with this weaving Say, the other examples don't have to be repeated. ^ f method (as an example in Figure 15) 'expanded graphics by merging additional processing units for holographic transformation and encoding The system's 3D rendering pipeline (10) method is the one for the present invention. The so-called holographic conversion and encoding processing unit is used here. This name is substituted. The complete pipeline is arranged directly downstream of the 3D graphics pipeline. Each cluster; ., &amp;, and fountain (PipeHne) data is transmitted to the corresponding cluster on the display; from now on, the following narrative will focus on - A single-cluster level display design. - A z-axis corresponds to the buffer and the color corresponds to the buffer (color dance should be R, color should be G, color miscellaneous ^ handle handle interface between the official line. Please refer to the simplified diagram in Fig. 15. The 'z-axis corresponding to each of the individual points in the pixel coordinates contains a z-value, which is scaled and can be expressed in various, monthly levels. 2 The typical scale of the value is in the range between 〇Μ 1.G 'but it may be other Fan ®. Its level of clarity will be based on the number of bits, ie, usually 8, 16, or 24 bit. In modern graphics subsystems, color The resolution should be 24 bit, that is, each bit P, 'work', 彔, blue) 8 bits each. The color corresponds to the frame buffer of the frame 45 200916986 =, the buffer &amp; On the surface, these two buffers (including the corresponding data of the z-pull = material and color) will be defined to constitute (10) the coloring pipeline is finely added __ and the fully recorded _ interface. z (10) should be targeted - The display wavelengths are provided, but R, G, and B have no specific wavelengths. The copy legs of the corresponding data are provided separately for the other wavelengths. 'A holographic image can only be specific to _Showing silk can be calculated. This is why it must be _ every object point _ three calculations _, that is, for each primary color money set - person 'red (10)), green aG), and blue (λ Β). Other colors can be created with these three colors, and this color can be adjusted in a sequential or synchronized manner. In order to speed up the processing, it uses at least two additional full pipelines so that the holographic image calculation can be sub-defined so that it can get the results of all three separations simultaneously. To do this, it must read the corresponding data from the Ζ axis to the additional memory segment deletion and simplification (please refer to (5), these regions can be accessed independently. Therefore, this can prevent operations involving the collision segment. (such as the axis corresponding to the material) can cause obstacles to each other. Therefore, each memory segment should be separated in theory. The color correspondence data 关于 in the color G and B _ capacity will also be copied to the color The secret memory segment corresponding to G and color corresponds to ensure that the three color separations can be accessed independently (see _ 15). Similarly, each memory segment may also be physically separated to prevent it from being recorded in It. Synchronization access problems such as maphore, mutual exclusion (ie, "muscles"), etc., which occur during the access period and which reduce or even eliminate the miscellaneous display design, may be related to system performance. There are adverse effects. However, although the memory segments may actually be separated from each other', they should preferably be located in the same cluster range of the display. Note that a number 46 200916986 is a dependent variable ( Or reduced the touch type) A typical method used to restrict access to shared resources (eg, storage space) in a lexical operating environment; mutexes are used in parallel programming to call key segments through fragments of computer code. The way to avoid using a shared resource at the same time, for example, a whole variable. It is assumed below that a hologram image is composed of multiple sub-images, and the first holographic image is represented by a lens. This lens can be described by a lens function Y (a) to U (10)). The constant G, including the focal length of the lens (1), is calculated before applying the lens function so that the value of f can be used in all three pipelines. The value of f is therefore related to the color: 峨 (一) 贞峨, the metaphor of 彡 脱. This can have the advantage of a lens function _ because the guzzoscope is symmetrical on its k-plane. In order to describe a lens in detail, this function only needs to be applied to a quadrant. The value of the lens function calculated in one quadrant can then be applied to the three quadrants of the basin using the symmetry principle of sign. '' &amp; also varies depending on the wavelength, the wavelengths of the three primary colors (R, G, B) are essentially different. The H # value does not need to be calculated because, because of the fact that it is used - a silk Laser or ~ originally produced a mother-wavelength, so the wavelength is red; however, the value should be substituted in the calculation to calculate the Ct of each primary color of the display (see Figure 15). According to the collected objects, Shan Zaiqi #咖崎崎 uses a prism function (see _ 15) to change the direction of light transmission. In the prism function, a constant also includes the wavelength λ. This f-number will also have some. Because the three-colors have different wavelengths, the value of the sigh is in the three-year-old pipeline. Each tube has a specified value of 47 200916986. The 'lens function and the 稜鏡 function are now subjected to complex multiplications 15 〇 3, 15 〇 4, and 1505 as shown in FIG. Next, the random phases 1506, 1507, and 1508 are applied to the results of the complex multiplication of the lens and the prism function. The goal of this method is to avoid producing peaks of brightness or bright spots on the observer plane. The color of the job is then adapted to the light intensity of the dragon to modulate the individual hologram images 1509, 1510, and 1511. In the next step, this sub-image will perform a complex addition to form a total holographic image for the cluster (see Figure 15). Now, the results obtained can be used for subsequent shouting (if appropriate) using the addition in the hologram display cluster, for example, applying a correction corresponding or grayscale image (gray_positive), since this will only depend on Μ The green nature, so it is best to make corrections in _ paragraph. Next is the encoding process. The holographic image can be reconstructed with a variety of colors. The encoding algorithm (see Figure 15) can vary greatly depending on the Μ used, k can be encoded according to her encoding, silk encoding, or another way. Those skilled in the art will appreciate that some aspects of the display design described in this section are disclosed in more detail elsewhere in this application.

F.可以以一個全像計算管線透過擴充顯示卡的3D管線(Pipeline)fG 維空間中的各個點進行循序全像轉換並在同一基板上進行計算的全像影像 顯示器 則面A項的齡設計可以湘先前的專_方法實絲進行空間光 48 200916986 調制器的編碼計算。此外’前面A項的顯示設計可以糊—個可在執行全 像計算時減少制祕的綠來實施。魏可錄行全料算時減少時間 延遲的方法的-個細㈣下麻，㈣於衫熟悉賴猶的人來說， 其他的範例自不必多加贅述。 對-個可在像素近旁進行計算的全像影像顯示器來說，這個顯示設叶 的-個目標是，減少時間延遲⑽於其他的全像計算來說）。這將會因為 用於即時全像轉換和編碼的額外《模_導致（例如）現行使用的顯示 卡（3D管線(Pipeiine))在架構上的擴大。 -般而言’在執行-個全像轉換計算之前，整個三維景象已經透過數 次的邪轉換和光度計算來構成。構成景象中的物件的各個基元（例如， 點、線、三角）會在3D處理的管線末尾時進行像素化。整個結果接著會 放在兩個記憶段供铜。這兩個區段是包含有觀察者所相的景象的 色彩值«料）的影格緩衝區、和—個包含有從觀察者的位置看 會以-個尺度比例來表現景象的緩衝區對應龍的z軸緩衝區。在先前的 專利的方法巾，全像轉換㈣碼辩对在私躺結果（_記憶體區 段）可供制時才會開始，因為這兩個記憶體區段的存取必須如此。而、言 會導致一個視訊框格產生一個眩pg 這 固時間延遲。這樣的時間延遲在某些互動 應用（例如遊戲裝置）中可能A 4 , ' 來重大的影響。如果延遲時間太長 供操作者猶·敝應日麵㈣錢社短，而使縣料壯法齡 某些原本可能可以執行的動作。—個影格的延遲時間（在6Q Hz的顯= 置中不a ’me Π ms) ’ f悔高速的遊戲來說可能會帶來重大的影響。: 49 200916986 全像顯示“只有在有這方關顧可雜之下才有被料接受的可能 (像是電子視訊賴的玩家這_目標鱗是絕對不能忽略的）。 三維立體全像絲技術在軍事制上也有它的優點，像是可以透過三 維的視野來觀察敵軍、或者其储訊（如地形魏），這可以改善遂行戰鬥 的有效性’遠較二維資料顯示器為佳1如果將這個顯示雜祕戰鬥任 務中的軍事用途上，上述的時_遲可能會導致執行任務的人員傷亡、或 導致昂貴軍事設備損壞或毁壞的嚴重後果。引此，減少上述的時間延遲將 可以改Q —維立體全像成像技術在軍事上的應帛的有效性。 為了減少這個延遲時間，我們並不須要等候全部的色彩及z轴緩衝區 對應資料都可以使用才能開始處理。相反的，全像計算將會在經過邪管 線(Pipeline)處理之後、只要空射已經有—個點可用的情況下立即進 行。因此，我們可以說這個3D管線(pipeline)可以透過一個全像管線來 擴充。 用於進行全像轉換和編碼的計算時間最好應不超過3D管線 (Pipeline)中進行一個3D點的計算所需的時間，因為若非這樣，將會造 成進一步_鋼延遲。這個概念在次全像影像層面上已經可以輕易實現， 因為在k種If况下只有必要的資訊片段需要進行處理。為了說明這一點， 我們假口又如果將全像轉換從空間中的一個單一的邪點套用到一個全像影 像或M的整體規模’可能會產生大約1，咖倍以上的額外運算負荷量: 如此-來，使用現行可_運算硬齡進行㈣計算將可能會無法實現。 圖8顯示出―個次全像影像的概紛口它的相關說明。圖18則以圖解方式 200916986 解說了在本發’―她*設計的這讎射所偏好細敎全像影像。 由於次全像影像比SLM更小，因此每一個次全像影像的計算都會比一麵 跨整個SLM的完整全像影像的計算更為快速。除此之外，各個次全像影像 也可以依序計算’相較於計算-娜跨 SLM的完整全像影像的情況 (這只有在接收到-個完整的影像龍框格射能執行），這可以大幅減少 時間延遲。在比較兩個附圖⑽和應）時，可以清楚的看出如果田採用 次全像影像，計算每—個物件點的運算錢量將會大幅降低，料相較於 整個SLM來說，-個次全像雜巾的像雜數量要少得多。 在本發明的一個顯示設計的某些範例中，位置最靠近觀察者的點的次 全像影像（® 16)會儲存在—個次全像影像緩_巾。胁每 管線叫el㈣資料會傳送給顯示器上相應的叢集（㈣）；從現在起以下 的敘述將著重在-個單—叢集層級的顯示設計h關於v⑽的大小以及 vow的方向和與SLM之間的距離的資料會供應給叢集做為輸入來進行叶 ^圖Π)。顯示器的每-嶋都有它自己的檢視表咖存它所顯示的 夂全像影像的編碼’這可能是—個或多個次全像影像。每當有—個更靠近 贿者的_誠赠，就會進行對應於這個_次全像影像（㈤的叶 异U參閱圖π)，也就是說’全像轉換在次全像影像的尺寸已經確定後才 2行。織，SLM的叢集的内容並不能只是簡單的以次全像影像來覆寫， 因為—個SLM像素格可能包含有來自多個次全像影像的資訊。這也正它會 針對一個在位置xy輸入的次全像影傻 曰 像（SHn-1)(這個次全像影像在當時也F. It is possible to use a holographic calculation pipeline to perform sequential holographic conversion by expanding all points in the 3D pipeline (Pipeline) fG dimension space of the display card and calculating on the same substrate. The encoding of the spatial light 48 200916986 modulator can be performed by the previous special method. In addition, the display design of the previous item A can be implemented by a green color that can reduce the secret of the system when performing the hologram calculation. Wei Ke recorded all the time to reduce the time delay method - a fine (four) under the hemp, (four) in the shirt familiar with the people of the lie, other examples do not have to repeat. For a holographic image display that can be calculated near the pixel, the goal of this display is to reduce the time delay (10) for other holographic calculations. This will be due to the architectural expansion of the additional "mode" resulting in, for example, the currently used display card (Pipeiine) for instant omni-directional conversion and encoding. - Generally speaking, the entire three-dimensional scene has been constructed through several evil conversions and luminosity calculations before performing the hologram conversion calculation. The individual primitives (eg, points, lines, triangles) that make up the object in the scene are pixelated at the end of the 3D processed pipeline. The entire result is then placed in two memory segments for copper. These two sections are the image buffers containing the color values of the viewer's scene, and the buffers containing the buffers that represent the scene in a scale from the observer's position. The z-axis buffer. In the method of the prior patent, the holographic conversion (4) code discrimination is only started when the private result (_memory segment) is available, since access to the two memory segments is necessary. However, the word will cause a video frame to produce a glare pg, which is a fixed time delay. Such time delays may have a significant impact on certain interactive applications (such as gaming devices). If the delay time is too long, the operator should wait for the sun (4) Qianshe is short, and the county is expected to be able to perform some actions. The delay time of a frame (not a ’me Π ms in the 6Q Hz display) ’ f regrets the high speed game may have a significant impact. : 49 200916986 The hologram shows that "only if there is such a concern, there is a possibility of being accepted (such as the player of electronic video game, the target scale is absolutely not negligible). Three-dimensional full-image silk technology It also has its advantages in military system, such as the ability to observe the enemy through three-dimensional field of view, or its storage (such as terrain Wei), which can improve the effectiveness of the battle of the limp' is far better than the two-dimensional data display. In the military use of this display of the mysterious combat mission, the above-mentioned time delay may result in the death or death of the person performing the mission or the serious consequences of the damage or destruction of the expensive military equipment. Therefore, reducing the above time delay will be possible. Q-dimensional stereo holographic imaging technology is effective in military response. In order to reduce this delay time, we do not have to wait for all the color and z-axis buffer corresponding data can be used to start processing. On the contrary, all Like calculations will be done immediately after the Pipeline process, as long as the airshots already have a point available. So we can say This 3D pipeline can be augmented through a hologram pipeline. The calculation time for hologram conversion and encoding should preferably not exceed the time required for a 3D point calculation in the 3D pipeline (Pipeline). This will cause further _ steel delay. This concept can be easily implemented at the sub-holographic image level, because only necessary information fragments need to be processed in the case of k. In order to illustrate this, we have a fake Applying a holographic transformation from a single evil point in space to a holographic image or the overall size of M' may yield an additional computational load of approximately 1 or more than the coffee: So - come, use the current _ operation hard The age calculation (4) calculation may not be possible. Figure 8 shows the relevant description of the holistic hologram image. Figure 18 illustrates the 雠 她 她 她 她 她 在 她 她 她 她 她 她 她 她 ― ― ― ― ― ― ― ― ― ― ― ― ― ― ― The preferred holographic image is preferred. Since the sub-holographic image is smaller than the SLM, each sub-image image is calculated more than a full hologram image across the entire SLM. In addition, in addition, each holographic image can also be calculated in sequence as compared to the calculation of the complete holographic image of the S-SLM (this is only received - a complete image of the dragon frame grid energy Execution), this can greatly reduce the time delay. When comparing the two figures (10) and should be), it can be clearly seen that if the field uses sub-holographic images, the calculation of the amount of money for each object point will be greatly reduced. The number of imaginary linings is much less than that of the entire SLM. In some examples of a display design of the present invention, the sub-image of the point closest to the observer is imaged. (® 16) will be stored in a sub-holographic image buffer. The data for each line called el (four) will be transmitted to the corresponding cluster on the display ((4)); from now on, the following description will focus on the - single-cluster level The display design h about the size of v (10) and the direction of the vow and the distance between the SLM and the SLM will be supplied to the cluster as input for the leaf ^ map). Each display of the display has its own view table to store the code of the holographic image displayed by it. This may be one or more sub-holographic images. Whenever there is a _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Only 2 lines have been determined. The content of the SLM cluster cannot be simply overwritten with sub-holographic images, because an SLM pixel may contain information from multiple sub-images. This is also true for a sub-image stupid image (SHn-1) input at position xy (this sub-image is also at the time)

_顯示在Μ的叢集上）搜尋-個檢視表的原因。在讀取來自咖的SH 51 200916986 的内容之後，它會計算出目前顯示的SH (I)與新的紐⑽）之間的 差（請參閱圖17)。 在空間中有一個3D點的情況下，如果這個點比先前的-個點更靠近 觀察者，則會繩嫩1料彳爾，肌t離到LUT中 取代舊的亂1 (請參閱圖17)。現在，兩者之間的差SHD會被加到SLM中 的值上’並且儲翻-悔彡格緩彳祕中。独過程之後緊接著就是進行編 碼以及可能的修正（請參閱圖17)。 衣置（SIi〇會提供它的配置資訊（例如類別解析度）給 （1閱圖17)’&amp;表不它將可以連接任何種類的全像顯示裝置 (SLM)。這練置可能林_大小、像餘數量、或甚至是編碼類別。 因此’這個解決方案並不會限定止能使用—個特定類別的 G. 了在同—基板上執行計算的全像影像顯示器，全像顯示器並可隨機編址 則面A項的齡設計可㈣贱前的翻的方法實絲進行空間光 _器的編碼計算。此外’前面A項的顯示設計也可以利用一個可提供更 佳，的程序來執行全像計算的方法來實施。町將針對—個可提供更佳的程 序來執仃全像計算的方法的_個範例加以制，但對於許多熟悉這類技術 的人來說，其他的範例自不必多加贅述。 這個顯示設計的-個目標是，透過在應用中_次全像影像的特性上 的優點來減少必須從容產生模組（例如，顯示卡）傳送到視覺化模 52 200916986 組（也就是全像顯示器）的資料量。 在先前的專利中，從内容產生單元（例如，顯示卡）到視覺化模組（例 如，LCD或陰極射線管（CRT)監視器）所進行的影像資料傳輸，會傳送一 個影像的整個内容（從上到下、一條掃描線接一條掃描線），就像傳統的射 線管監視器一樣。在高清晰度電視（HDTV)方面，解析度可高達384〇 χ 24〇〇 像素（IBM(RTM)Berta顯示器現在的ΠΙΑΜΑ等，請參閱網站說明， 例如：http://www.pCmag.com/article2/0，1895,2〇38797 〇〇 asp),這並 不會有問題，因為所需的資料量可以透過各種標準化的介面，像是數位視 覺介面（DVI)或高清晰度多媒體介面（HDMI)，以足夠的高速傳送。 不過，理想的全像顯示裝置則會需要有遠比此更高的像素數目才能在 觀察者平面中產生一個大小為一到幾cm寬的虛擬觀察者視窗（v〇w)(相 反的，若是較原始的裝置，則大約只有5刪寬）。較大的v〇w有非常多 的好處，因為就商業使用中的可靠性來說，如果這個視窗越大，這個全像 顯不裝置也越可靠。這是因為對於具有追蹤功能的全像顯示器（像是追蹤 系統或位置瞄準裝置，用來追蹤觀看者的眼睛相對於顯示器的位置）中的 其他零組件的要求，在這種情況下將會比較低。此外，如果裝置不具有追 蹤功能’若是將VGW的大小增大-些’也可以改善觀看者頭部微幅移動的 容忍度。 這個顯示設計的一個目標是，透過全部或至少一部份的全像計算會在 像素矩陣断來減少必織魄產賴組傳制全細示$巾的視覺化模 組的資料量。 53 200916986 在前面所述的先前專利的資料傳輸過程中，它會傳輸所有的資訊包 括那些從-個框格到下-個框格都沒有改變的資訊片段。因為—個全像影 像會在—個三維空間中重建影像的各個點，因此只要知道哪些點相較於^ 前的框格來說有所改變就已經足夠’在後續的處理中也只會考慮這些點（請 參閱圖19)。 ^ 一個單一的物件點由一個次全像影傻ς 丨u人王像々像mi產生，它的大小則取決於觀 _的位置。由於-個SLM像素格可能不只包含—個次全像影像的資訊， 還可能包含多個次全像影像的資訊，因此 J Μ彳异在位置xyz上的舊的點 的SH和在同樣的位置xyz上的新的點的sk之間的差別。在本發明的一 個顯示設計的-纖财，這健Μ全像騎shd歸可邮μ上 重新進行編碼。 在顯示器内部或外部的電路組件會接收3d影像資料，這個資料包括 =每—個框㈣—购_賴㈣-们滅衝區對應資 =會計_的赌之_差，卿巾的_卿。之後，更新了 細影像差„料的形式傳送給顯示器的全像轉換單元。如圖 母個王像轉換單元都會傳送相對於重建點或用來在SLM上 進行編碼的點的3D差里赴旦y 十 干次 ‘讀感。如果—彳嚼定叢集的連續祕的顯 傳送資料給全像轉換單元…了^政而可以忽略的差異，則並不須要 ㈣田# 這可以加速顯示系統的有效SLM更新速率。系 統中用來產生SHD的部 又w十系 °以稱之為内容產生模組’，，它可能包括運算 力月b和一個顯示卡。接基 再將這些次全像影像傳送給每一個叢集。叢集所 54 200916986 執仃的第紅作是處理所接收到的資訊，將全像影像資料與有關，的 大小和位置的資料分開。叢集的工作還包括將删寫入到適當的臟像 素格中，使SH可以正確的顯示在適當的slm位置上並且具有正確的大 /J、〇 除了认王像衫像SHd (或者是一個新的框格的SH)之外，次全像影像 以像素數表示的大小以及它在顯4叢針的㈣也可以指定。在全像顯 不叢集中（如圖2(3巾的範騎示）有—個分解器'，它會將計算出來的全像 影像顯示·分簡次全像影像㈣叹大小和位置#訊。後面這兩個值 的目的是用來計算次全像影像在麵中的位址範圍，使次全像影像紐或 shd的資料可以被寫入到叢集中的正確Μ像素格上。 常見的SLM是主動矩陣式顯示器，它的像素格必須持續更新才不會讓 資訊流失。如果只有新的内容會被寫入到Μ上，則其他區域中的資訊將 會流失（例如，請參閱圖19 :其中的四個黑色圓點將不會再出現基於這 個理由’在這種情況下可以使用—個特殊的隨機存取記憶體（麵）來使輸 入侧只會寫入新的SH或SH。，而在輸出側則會完整的讀取整個記憶體並將 完整的資訊寫入到SLM上。雙璋_或其他允許同時讀取和寫入的記憶 體系統所_發_作用（如以上所述）’即可以細在這樣的目的上。 根據3D景象中的變化情況’將會傳送哪些點的資訊，這會在内容產 生單兀中決定。0此’減少餘流_作會在資料被傳送到全像顯示裝置 之前就會執行。這些資訊可以以任何轉傳輸，因為次全像影像會有額外 的資訊加以補充，如以上所述。這大致上是掃贿之_差異，就像先前 55 200916986 的專利的視覺化系統中的資料傳輸一樣。 在用戶側，也就是在内容已經產生後，資料是否要傳送的一個決定會 在資料開始傳送之前先行決定’如本發明輸示設計巾所述。如果内容已 故完全改變，像是在-個場景轉換之後或者錢示—個完全不___樣的景象 的情況下’將會有非常多的對應於各個3D物件點的次全像影像必須傳 送。典型上來說’細可以斷言：—個SLM的腑度越高，以傳送次全像 影像來取代傳送整個全像影像的優點也越明顯。 H.在像素空間中具備有運算功能的顯示器 在本發明的-個顯示設計的—個進一步的範例中，有一個顯示器會用 來顯示影«料，這可能是-般_示資料或者也可能是已經根據光強度 對應資料和緩衝區對應資料計算出來的全像影像顯示資料。先前的專利中 的顯示器在先天上的_是，它們g要不是與顯示電路轉佈置在同一基 板上的電路組件。這些額外的组件必_置在顯示器基板之外的—個 單獨的基板上。這會導致—些不利的㈣，像是較大魏置體積和重量。 但消費者雜追求的卻是更輕、薄、短、小的顯示裂置。如果採用 段，像是圖25中所示的手段，這些像是裝置體積較大和重量較重之類的門 以迎刀而解。如果將運算單元佈置在靠近顯示器的各個像素的地 示計二在顯不任何時的時間延遲（這已經由運算單元針對顯 合册來、域的時間延遲的減少對於諸如高逮遊戲裝置的應用來說將 的好處，或者在用於軍事用途的裝置中 … 度也可能帶來軍事上的優勢。 更么的裝置性能和速 在圖25所不的顯不器中，運算功能會在顯示器的各個叢集處，在顯示 56 200916986 ―各個顯示像素之間、或者在顯示器的各個顯示像素旁_執行。用 來執行運算功能的空間是在與顯示_板所在的同—基板上。在圖烈 ^於運馳謂#輪賴魏。嶋纖術的人來 5兒，其他的範例自不必多加贅述。 吸收 在電腦_巾，“魏（㈣㈣⑹”職驗科綱—個較靠近 視平面的物件遮蔽（或吸收）一錄遠離視平面的物件的方式。在沙顯 示器的圖形管財’人們會採[觀收剔除的形式在進行騎和試映之 前先行移除㈣的表面。而在全像的魏巾，“魏，，賴示設計會牵涉 到必須確保轉虛峨察者職較近的物件點麵蔽沿著蝴的視線上距 離虛擬觀察者視窗較遠物件點。 我們對—個全像顯示器所希望的吸收行為的-個範例如圖29所示。在 圖29中，從所示的眼睛位置，應該不可能看得到立方體的厚度側，因為這 一面被立讀鈴魏看者的—面魏了。如果VGW的大付眼睛瞳孔大 小的數倍之大’則觀看者將可以從不同的方向看這個立方體，以便能夠看 到這個立方體的厚度側。但如果是—_單的吸《阮設計，這個立方體 的厚度側將不會在SLM上進行編瑪，所以即使觀看者改變觀看方向，觀看 者還是看不到立方體的厚賴，®為它並沒有在SLM上進行編碼。 57 200916986 在圖30中，觀看者從—他 5 ;圖29中所示的方向來看這個立方體 以便此夠相対體的厚相 〇n 彳―如果疋一個簡單的吸收顯示設計，如果 ςίΜ 文处理，這個立方體的厚度側也不會在 SLM上進仃編碼，所以圖3〇中的 ... 看者遏疋無法看見立方體的厚度侧，因 為匕〜又有在SLM上進行編竭： 虿針對圖29中的立方體的厚度側 所屬的物件點進行重建，因此在 ^卯也沒有這個立方體的厚度側經過重建 的物件點。 圖30中所示的問題的一個解 、 解决方法是將簡分成兩個或多個區段， 然後針對每一個V(观區段來#i ^ &amp; 奴來重建各個物件點。每—個區段的大小最 好大致和人類眼睛的瞳孔大小相當。 在圖31中，從眼睛位置1，顴 觀看者將可以看到物件點1但不能看到 被遮蔽的物件點2。而從眼睛位署9 ^ &lt; 月位置2 ’觀看者將會看到物件點2，但看不 到從這個位置和觀看方向所無法看 ^ J初件點1。因此，從眼睛位置2，_ is displayed on the cluster of )) the reason for searching - a view. After reading the contents of SH 51 200916986 from the coffee, it calculates the difference between the currently displayed SH (I) and the new New Zealand (10) (see Figure 17). In the case of a 3D point in space, if this point is closer to the observer than the previous - point, then the rope will be tender, and the muscle will leave the LUT to replace the old chaos 1 (see Figure 17). ). Now, the difference SHD between the two will be added to the value in the SLM' and the memory will be saved. The unique process is followed by coding and possible corrections (see Figure 17). Clothing (SIi〇 will provide its configuration information (such as category resolution) to (1 see Figure 17) '&amp; table will not be able to connect any kind of holographic display device (SLM). This practice may be forest _ Size, number of images, or even coding categories. So 'this solution is not limited to use - a specific category of G. A holographic image display that performs calculations on the same substrate, a holographic display and The random design can be used to perform the calculation of the spatial light_device. The holographic calculation method is implemented. The town will work on a method that provides a better procedure for performing holographic calculations, but for many people familiar with this technology, other examples are It is not necessary to repeat this. The goal of this display design is to reduce the need to passively generate modules (eg, display cards) to the Visualization Module 52 200916986 group by virtue of the characteristics of the _ holographic image in the application ( The amount of data in the holographic display. In the prior patents, image data from a content generation unit (eg, a display card) to a visualization module (eg, an LCD or cathode ray tube (CRT) monitor) Transmission, which transmits the entire content of an image (from top to bottom, one scan line and one scan line), just like a traditional ray tube monitor. In high definition television (HDTV), the resolution can be as high as 384 〇. χ 24 〇〇 pixels (IBM (RTM) Berta display, etc., please refer to the website description, for example: http://www.pCmag.com/article2/0, 1895, 2〇38797 〇〇asp), this and There is no problem because the amount of data required can be transmitted at a high speed through a variety of standardized interfaces such as Digital Visual Interface (DVI) or High Definition Multimedia Interface (HDMI). However, an ideal hologram display device It would be necessary to have a much higher number of pixels to create a virtual observer window (v〇w) that is one to several centimeters wide in the viewer plane (in contrast, if it is a more primitive device, then only 5 widening.) Larger v〇w has a lot of benefits, because in terms of reliability in commercial use, if the window is larger, the holographic device is more reliable. This is because there is tracking The requirement for a functional hologram display (such as a tracking system or position aiming device to track the position of the viewer's eye relative to the display) would be lower in this case. In addition, if the device Without the tracking function 'If you increase the size of the VGW - you can also improve the tolerance of the viewer's head's slight movement. One goal of this display design is that all or at least part of the holographic calculation will be The pixel matrix is broken to reduce the amount of data that must be woven by the group to visualize the module. 53 200916986 In the data transfer process of the prior patent described above, it transmits all the information including those pieces of information that have not changed from one sash to the next. Because a hologram image reconstructs the various points of the image in a three-dimensional space, it is sufficient to know which points are changed compared to the sash before ^, and will only be considered in subsequent processing. These points (see Figure 19). ^ A single object point is produced by a sub-image ς 人u person like a mi, its size depends on the position of the view _. Since the -SLM pixel may contain not only the information of the sub-holographic image, but also the information of the multiple sub-images, J is different from the SH of the old point on the position xyz and in the same position. The difference between the sk of the new point on xyz. In one of the display designs of the present invention, the gimmick is re-encoded on the shod. The circuit components inside or outside the display will receive 3d image data. This data includes = every box (four) - purchase _ _ (four) - we rush area corresponding capital = accounting _ gambling _ poor, qing _ _ Qing. After that, the holographic conversion unit that transmits the fine image difference to the display is updated. The mother image conversion unit transmits the 3D difference with respect to the reconstruction point or the point used for encoding on the SLM. y ten times 'reading sensation. If - 彳 定 定 的 的 的 的 的 的 的 丛 给 丛 丛 丛 丛 丛 丛 全 全 全 全 全 全 全 全 全 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ SLM update rate. The part used to generate the SHD in the system is called the content generation module', which may include the computing power month b and a display card. The base then transmits these sub-images. For each cluster. Clusters 2009 2009986 The first red book is to process the received information, separating the holographic image data from the size and location of the data. The work of the cluster also includes writing the deletion to the appropriate In the dirty pixel grid, the SH can be correctly displayed at the appropriate slm position and has the correct large /J, in addition to the avatar like SHd (or a new sash SH), the second full Like image to image The size of the number and the (4) of the 4 cluster needles can also be specified. In the whole image display cluster (as shown in Figure 2 (3), there is a resolver, which will calculate the full Image display · Sub-simple holographic image (4) Sigh size and position #. The purpose of these two values is to calculate the address range of the sub-holographic image in the face, so that the sub-holographic image or shd Data can be written to the correct pixel grid in the cluster. A common SLM is an active matrix display whose pixels must be continuously updated so that no information is lost. If only new content is written to it. , the information in other areas will be lost (for example, see Figure 19: four of the black dots will no longer appear for this reason 'in this case can be used - a special random access memory (face) to make the input side only write a new SH or SH. On the output side, the entire memory will be read completely and the complete information will be written to the SLM. Double 璋 or other allow simultaneous reading Take and write the memory system Said) 'that can be fine for such a purpose. According to the changes in the 3D scene', which points of information will be transmitted, this will be decided in the content generation unit. 0 This 'reduction of the residual stream _ will be transmitted in the data It will be executed before the holographic display device. This information can be transferred in any way, because the sub-holographic image will be supplemented with additional information, as described above. This is roughly the difference between the two. The data transmission in the visualization system of the patent of 200916986 is the same. On the user side, that is, after the content has been generated, a decision on whether the material is to be transmitted will be decided before the data is transmitted. 'As described in the design of the present invention. If the content has completely changed, such as after a scene change or a money--a scene that is completely ___-like, there will be a lot of sub-holographic images corresponding to the points of each 3D object. Must be transmitted. Typically, the fine can be asserted: the higher the sensitivity of an SLM, the more obvious the advantage of transmitting a sub-image is instead of transmitting the entire hologram. H. Display with Arithmetic Function in Pixel Space In a further example of the display design of the present invention, a display may be used to display the image, which may be a general information or may be It is a holographic image display data that has been calculated based on the light intensity corresponding data and the buffer corresponding data. The displays in the prior patents are innate _ is that they are not circuit components that are arranged on the same substrate as the display circuit. These additional components must be placed on a separate substrate outside the display substrate. This can lead to some disadvantages (four), such as larger volume and weight. However, consumers are pursuing a lighter, thinner, shorter, and smaller display. If a segment is used, such as the one shown in Figure 25, these are like the larger and heavier the door of the device. If the arithmetic unit is arranged close to the respective pixels of the display, the time delay of the display 2 is displayed at any time (this has been caused by the arithmetic unit for the reduction of the time delay of the domain, for applications such as high catching game devices) For the benefit, or in the device for military use... Degree may also bring military advantages. More device performance and speed in the display of Figure 25, the computing function will be on the display Each cluster is executed on display 56 200916986 - between each display pixel or next to each display pixel of the display. The space used to perform the computational function is on the same substrate as the display_board. Yu Yunchi said that #轮赖魏. The people of 嶋纤术 came to 5 children, other examples do not need to be repeated. Absorbed in the computer _ towel, "Wei ((4) (4) (6)") (or absorb) the way in which objects away from the plane of view are recorded. The graphics in the sand monitors 'people' will pick up the surface in the form of pick-and-remove before removing and simulating (four) In the hologram of the Wei towel, "Wei,, the design of the display will involve ensuring that the object closer to the imaginary position is covered by the distance from the virtual observer window along the line of the butterfly. We An example of the desired absorption behavior for a hologram display is shown in Figure 29. In Figure 29, from the eye position shown, it should not be possible to see the thickness side of the cube because this side is being read. Wei's face is super-faced. If the VGW's big eyes are several times larger than the size of the pupil, then the viewer will be able to see the cube from different directions so that the thickness side of the cube can be seen. But if it is - _ single suction "阮 design, the thickness side of this cube will not be programmed on the SLM, so even if the viewer changes the viewing direction, the viewer can not see the cube's thick, ® is not on the SLM In the case of Figure 30, the viewer looks at the cube from the direction shown in -5; Figure 29 so that the thickness of the cube is 〇n 彳 - if a simple absorption display design, if ςί According to the text processing, the thickness side of this cube will not be encoded on the SLM, so in Figure 3... The viewer can't see the thickness side of the cube, because 匕~ is also compiled on the SLM. : 重建 Reconstruct the object point to which the thickness side of the cube in Fig. 29 belongs, so that there is no object point on the thickness side of the cube. The solution to the problem shown in Fig. 30 is to solve The simplification is divided into two or more sections, and then each object point is reconstructed for each V (viewing section #i ^ &amp; slaves. The size of each section is preferably approximately the same as the size of the pupil of the human eye. In Figure 31, from eye position 1, the viewer will be able to see the object point 1 but not the obscured object point 2. From the eye position 9 ^ &lt; month position 2 'viewer will see The object points 2, but you can't see it from this position and the viewing direction. So from the eye position 2,

觀看者可以看到在從眼睛位置丨顴I 硯看時會被物件點1遮蔽的物件點2。 物件點1和物件點2在次全像影傻〗 像〜像1和次全像影像2中分別都被吸收 了。 不過’在圖32中，從眼睛位置1 &amp; 置1和眼睛位置2都可以看見重合的物 件點1和物件點2,因為它們在次全像畢 王诼〜像1和次全像影像2中也都分 別重合。 、另外’吸收也可以在構成緩衝區對應資料和光強度對應資料的階段中 進行。在這麵沉下，最好針對每—個_ (也就是針鱗—個虛擬觀察 58 200916986 者視窗）各組織一〃 •個成對的緩衝區對應資料和光強度對應資料。 在本文件所包括的本發明的顯示設計的—個範例中，吸收會利用由佈 置在像素矩陣的空間中的電路組件所執行的計算來實施。這些電路 能包括ΤΠ。吸收也可以利用由佈置於和像素矩陣所在的同-基板上、作 在像素矩陣以外的空間中的電路組件所執行的計算來實施。 J.顯示卡的功能 -個圖形處理單元’或稱GPU (有時也稱為視覺處理單元或卿），是 -個個人電腦、工作站、或遊戲主機上的—個專用的_呈職置。現代 的GPU «縱及齡電_形方面料有料，_它們高度並行化的架 構讓它們在-系列複雜的運算上比典型的cpu ’、 現代的圖形處理單元_會將它們大部分_體用在盘3D電腦 圖形有_計算上。它_是絲加__取_的工作，如 紋理映射和呈現多邊形，後來有加上—些單元來加速幾何計算，例如將頂 點轉譯成不同細_嘴年來GPU方_發展包括支援可程式著色引 擎，它可以透續所支援的許多相同的操作、超取樣、和内插技術來操 縱頂點和紋理，以減少疊影，並具有極高精密度的色彩空間。 除了 3D的硬體之外，現今的Gpu也包括有基本的;；加速和影格緩 衝區的能力（通常採用-個視訊圖形陣列⑽相容模式）。另外，大部 分從廳年以後所生產的GPU也都支援丽色彩空間和硬體重叠（對 59 200916986 於數位影像播放非當 ‘‘動當直〜“要）財多在2_年以難造的GPU也支援有 rDm纟]、組（MPEG)基元，像是動作補償和反離散餘弦函數轉換 中近的顯示卡甚至可以在卡上直接進行高清晰度視訊的解碼，為 、处理料分攤了-部份的工作負荷。Yuv色彩空間的版本以—個光度 2個色衫分量的觀點來定義一個色彩㈣。娜色彩版本適用肌、 '和SECAM複合色彩視訊標準。 在本員申切所主張的全像顯示的環境下，顯示卡功能的設計牽涉到媒 保以增_能都可以在_轉進行全像物實施，而這裡所稱 的顯不器可以在像素矩陣所在的空間中執行所有的全像計算，或至少有一 勺王像。十算可以在像素矩陣所在的空間中執行。例如，這包括將著色 引擎設計成可以透過CPU所支援的許多相同的操作、超取樣、和内插技術 來繼·和紋理喊少疊影’並_極高精密度的色輕間，來加速需 要讀存取記憶體的工作’如紋理映射和呈現多邊形，以便加速幾何計算， 像疋將頂點轉譯為不同的座標系統，並執行牽涉到矩陣和向量操作的運 算:在計算全像影像方面，GPU高度並行化的架構讓它們在—系列複雜的 運算上比典型的CPU更為有效。此外，根據上述構想，本案之全像顯示器 也可以是不會在像素矩陣所在的空間中執行全像計算的—個顯示器。 、在本項申請所主張的全像顯示的環境下，顯示卡功能的設計也可能牽 涉到採用-個由佈置在像素矩陣所在的空間中、或在像素矩陣所在的空間 之外但在與像素矩陣所在的同一基板上的TFT所執行的3d著色管線 (ReridenngPlpehne)。以另一種角度來說，一個3d著色管線(触虹細 200916986The viewer can see the object point 2 that would be obscured by the object point 1 when viewed from the eye position 丨颧I 砚. The object point 1 and the object point 2 are respectively absorbed in the sub-image phantom image ~ image 1 and the second hologram image 2, respectively. However, in Fig. 32, the coincident object point 1 and the object point 2 can be seen from the eye position 1 &amp; 1 and the eye position 2, because they are in the second hologram 毕王诼~1 and sub hologram 2 They also coincide in each other. Further, the absorption may be performed in a stage in which the buffer corresponding data and the light intensity corresponding data are formed. In this case, it is best to organize each pair of _ (that is, the needle scale - a virtual observation 58 200916986 window) each pair of pairs of buffer corresponding data and light intensity corresponding data. In an example of the display design of the present invention included in this document, absorption is performed using calculations performed by circuit components disposed in the space of the pixel matrix. These circuits can include ΤΠ. Absorption can also be performed using calculations performed by circuit components disposed in a space other than the pixel matrix disposed on the same substrate as the pixel matrix. J. The function of the display card - a graphics processing unit, or GPU (sometimes called visual processing unit or qing), is a dedicated _presentation on a personal computer, workstation, or game console. Modern GPUs «Longitudinal and ageing _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ In the disk 3D computer graphics have _ calculations. It is the work of silk plus __ fetching _, such as texture mapping and rendering polygons, and later adding some elements to speed up geometric calculations, such as translating vertices into different s _ _ _ _ development including support for programmable coloring The engine, which traverses many of the same operations, oversampling, and interpolation techniques supported, manipulates vertices and textures to reduce aliasing and has a very high-precision color space. In addition to 3D hardware, today's Gpus also include basic; acceleration and frame buffering capabilities (usually with a video graphics array (10) compatibility mode). In addition, most of the GPUs produced after the year of the hall also support the Li color space and hardware overlap (for 59 200916986 in the digital video playback is not appropriate ''moving straight ~ "to") more than 2 years to make difficult The GPU also supports rDm纟], group (MPEG) primitives, such as motion compensation and inverse discrete cosine function conversion. The near-display card can even directly decode high-definition video on the card. - Part of the workload. The version of the Yuv color space defines a color (4) from the perspective of two luminosity components. The Na color version applies to the muscle, 'and SECAM composite color video standard. In the context of the claimed holographic display, the design of the display card function involves the media protection to increase the _ can be implemented in the ig-transformation, and the display device here can be executed in the space where the pixel matrix is located. All hologram calculations, or at least one spoonful of king images. Ten calculations can be performed in the space in which the pixel matrix is located. For example, this includes designing the shader engine to be able to pass many of the same operations supported by the CPU, overtaking And interpolation techniques to continue and texture shout less overlap 'and _ extremely high precision color light, to speed up the need to read access memory work 'such as texture mapping and rendering polygons, in order to accelerate geometric calculations, like转 Translate vertices into different coordinate systems and perform operations involving matrix and vector operations: GPUs' highly parallelized architecture makes them more efficient than typical CPUs in computing complex images. In addition, according to the above concept, the holographic display of the present invention may also be a display that does not perform holographic calculation in the space in which the pixel matrix is located. In the environment of the holographic display claimed in the application, the display card The design of the function may also involve the use of a 3d shading pipeline (ReridenngPlpehne) that is placed in the space in which the pixel matrix is located, or outside the space in which the pixel matrix is located but on the same substrate as the pixel matrix. In another way, a 3d coloring pipeline (touch rainbow 200916986

Pipeline)的魏，像是執行著色引擎的各種舰，將魏_專利 的顯示卡轉移到設置在液晶面板内的TFT上執行。 用 此外’根據上述構想，本宰之令德顧 尽茱之王L純可以是科在像素矩 在的空間中執行全像計算的—個顯示器。或者，同制，根據上述構想， 本案之全賴轉也可以是—財會在像素轉所麵”顿行全 算、但可以_存在於與像素矩陣所在的同—基板上的電路組件來二 像計算的顯示器。 i K. 2D □ 3D 轉換 在-個2D □ 3D轉換的範例中，來自—對立體影像的—個第一影像 和-個第二影像會傳送給顯示裝置，其中全部或至少一部份的全像計曾合 在像素所在的空間中執行，或者是在像素所在的基板上執行。2D □ 3D轉 換計算可以在像素轉所在的靖、或者是在像素所在的基板上的電路 組件上進行’或者可_來產錢规對應__麟應資料以 傳送給顯示器㈣路組件上進行，或者也可以在位於«聰技術的人都 清楚的其餘置上的電馳件上進行。第二個傳輸的影射能是這兩個立 體影像之關差聰像，_ —個差卿像通常將只會要求比—個完整的 影像更少的資料。如果是進行三維立體視訊顯示，第一影像本身可計算能 以現行影像與來自-_早的時關度的影像之間的差來表示。同樣的， 第二影像也可能以現行影像與來自—_早的時_度的影像之間的差來 61 200916986 表不。顯示裝置接著可能利用這個技術領域中用來在2D與三維立體（3D) y像之間進行轉換的已知計算程序，根據所接收到的資料計算出一個具有 相應的緩衝區對應資料的二維⑽影像。如果是—個彩色影像的情況， △曰而要有以二原、色呈現、並伴隨有各自對應的緩衝區對應資料的三個肋 刀色衫像。對應於2D影像和緩衝區對應的資料接著可能由用來顯示全像 ‘影像的裝置進行處理。這個裝置會在它的SLM上進行全像影像的編碼。為 充分有效的姻傳輸頻寬，在這個系統中傳輸的資料可能接受已知的舰 私序進行壓縮’並在顯示裂置上進行相應的解雜程序。 在其中執行2D □ 3D轉換的電路組件可能有權存取一個包含有一組 已知的3D形狀的資料庫，在這個資料庫中它可能會嘗試匹配它所計算出 來的3D資料；或者它可能有權存取—個包含有一組已知的那外形的資 料庫，在這個資料斜它可能會嘗試匹配輸入的2〇影像資料。如果與已 知的形狀之間發現有良好的匹配吻合，將可以加速計算過程，因為肋或 3D影像之後可能以一個相對的已知形狀來表示。3D形狀的資料庫可能會 轉有組（例如）運動明星（像是頂尖網球明星或足球明星）的諸如臉 P或身體的雑’以及歡運鱗地（像是知細球場或知名足球場）的 凡正或局部的形狀。例如，—個人的臉部的一個影像可能會以顯示裝 置所存取的-個影像、加上面部表情的變化（例如，可能是微笑或敵眉）、 再加上老如頭髮長度的某些變化（因為頭髮的長度在取得儲存的資料後可 ^ g長或’短）來表不。顯示裝置有權存取的資料如果出現有非常明顯 62 200916986 的差異使得它清趣示邮歸《所«麵的射化、_時，例如，— 個人的猶度已晴改變以及儲存物化經有—段报長的時 可以由顯示裝置進行更新。如果用於計算的電路組件碰到一個如或抑马 _在它所有權存取職料射找不到可以良好匹配吻合的形狀，則可^ 將這些新的形狀加入到記錄令。 加□邪影輸也可_—峨的、_顺相的心 像，利用這個技術領域中已知_於執行這類轉換的程序來執行。接著: 以將3D影像資料（緩衝區對應資料和色彩對應資料）傳送給顯示器來進 行全像計算並顯示出來。 次上U 3D轉換可能適用於將用來在一個全像顯示器上顯示的 貝料在乂個顯不為上全像計算會在佈置於像素矩陣所在的空間中的電路 、，且件上進订’或者至少有-部份的全像計算會在佈置於像素矩陣所在的空 間中、或在像素所在的基板上的其他位置的電雜件上進行。 L.視訊會談（3D Skype™) 根據第E3660065舰盟社群商標（EU c麵吻如加㈣）申請 案’我們知道Skype™可赠過―個㈣制祕提供嶋祕語音點對 點（卿）祕、以及齡分享、和即時訊息服務；也就是透過一個電腦 網路提供通訊»、_分享、Μ㈣訊息服務。 根據第Ε4521084舰盟社群商標⑽c〇{Mity如加fck)申請 63 200916986 案’我們域Skype™可他人提供冑腦贿和軟體騎，也就是用於 電信及網際網路語音協定⑽Ρ)應用、f料傳輸、及即時訊息服務的電 腦軟體及硬體設計；雜人建立及賴_ ;在—侧於全球化電腦網路 的電腦伺服紅控制他人的安裝及轉電腦軟體；提供可供線上暫 時使用、不可下載的電腦軟體以允許用戶使用v〇Ip通訊服務；提供線上 軟體供他人下載以允許用戶使用v〇IP通訊服務。 根據第2358090號英國商標（UK Trade Mark)，我們知道SkypeTM可 以提供網際網路存取、人π及快祕務；電信及電信服務；網際網路協定 (“IP”）服務；網際網路語音協（“VoIP”）服務；電子郵件及網際網 路通訊服務；經由第三方的電信服務；網際網路協定（“Ip”）對數值電 話號碼以及數值電話號碼對“IP”對應映對系統及資料庫；網域及網域資 料庫系統；出租電腦資料庫存取時間（由網際網路服務提供者提供）。 上述的任何服務都可以結合一個可以利用佈置在像素矩陣所在的空間 中的電路組件來執行所有全像計算、或至少可關關置在像素矩陣所在 的空間中的電路組件來執行—部份全像計算的全像顯示器來提供，除了 Skype™提供的VOip服務之外，還可以提供一個網際網路語音及全像影像 協定（VHIGIP)服務。在這種情況下，以上所述的程序可以由位於液晶面 板中的TFT綠行。此夕卜’上述的任何服務也可以結合一個不會在像素矩 陣所在的㈣巾執行全料冑的讀齡H來提供，除了咖,提供的 V0IP縣之外’還可以提供—個網際網路語音及全像影像協定（腿隱） 服務。此外，同樣的，上述的任何服務也可以結合—個不會在像素矩陣所 64 200916986 在的空間中執行全像計算、但會_位於與像素矩陣所在的同—基板上的 電路組件喊行錄計算的全侧示器來提供，除了咖？提供的贈 服務之外’還可以提供—個網際網路語音及全像影像狀（侧ιρ)服務。 此外，同樣的’上蘭任何服務也可以結合任何全像顯示錄提供，除了 yp提t、的νοίΡ服務之外，還可以提供—個網際網路語音及全像影像 協定（vhioip)服務。 此外，上述驗何服務也可以結合—個不會在像素矩陣所在的空間中 執行全像計算的全像顯示器來提供，除了提供的贈服務之 外，還可以提供-個網際網路語音及全像影像協定（v_⑻服務。 在以上所㈣服射，V_P可⑽網際網路語音城訊全像影像協 定（VVHI0IP)的形式來提供。腿或簡〇ιρ可以以即時或近即時的 方式提供，且這些網際娜龄可以允許在全都使财全像顯示器的兩個 人之間進行即時或近即時的全像視訊通訊。 M·編碼補償 在傳統攝影射，曝光纖是—細來針對可能料致呈現—個不戸 =像料侧素’來對_渐算妹的或計晝使用轉紐提供補償触 ^蝴靖崎爾㈣麵㈣贿、渡 2處理 '纖W繼。·_巾也爾轉門 4軟片速度的改變、或者其他的时來施加曝光補償。在攝影術中， 65 200916986 許多攝影機都會包括有這樣的功能以便讓使用者可以自動調整所計算的曝 光值。補償可以是分成多段的正補償（增加曝光）或者是負補償（減少曝 光）’通常每一段會增減1/3或1/2格光圈（f)設定，通常最多可以朝 增減（正負）方向作二或三段的調整。 就光學上來說，從鏡片的有效焦距的觀點來看，一個光學系統的f數 值表示瞳孔（細）開·直徑。在攝雜上，f數值通常會以多個分散 的格位來輕’稱為f設定。每—格“設定”都標示有相應的f數值， 且每-格都表示相當於前—格的—半的光強度。這相當独―個2的平方 根的倍率來減少瞳孔及光圈開孔的直徑，因此會使瞳孔（光圈）的面積減 半。 曝光補償會在當使用者知道攝影機所自動計算曝光值將會造成一個不 理想的曝光時採用。-個㈣亮_子縣的景象經常會造祕光不足， 而:個以陰暗_子為主的景象則經常會造成曝光過度。-個經驗豐富的 狀況下會有這種情況發生，並且知道要施加多少的補償 才月匕獲付有元吴曝光的相片。 ^述的任何功_可簡合_個可以在像素轉所麵同—基板上執 像計首3Γ、或至少可在與像素轉所在侧—絲上執行—部份全 十rrr的王像顯不考爽接征 L s σ /、。上述的任何功能也可以結合一個可以在傻去 矩陣所在的同—基板上執象素 空間中練—Α 冑王糾，、或至少可以在像素矩陣所在的 礼全像計算的全像顯示器來提供。此外，上述的任何功能 66 200916986 也可以結合任何全像齡ϋ來提供。顧可以在編碼麵巾或編碼步驟之 前施加到全像影像資料上，以提供—個可以看得更清楚的影像，也就是通 常觀察者會認為曝光正確的影像，而不會有曝光不足鱗光過度的問題。 Ν.眼睛追蹤 全像裝置可能具備有針對—個或多織看者的眼睛追縱功能。這對於 每個眼睛的觀看視錢尺推顿料尤其糊（例域喊幅只有幾毫 米的情況）。仙偏好制—條频料置_^追蹤步驟來使用者的 眼睛： 1) 透過偵測使用者的臉部來限定搜尋範圍 2) 透過偵測眼睛來限定追蹤範圍 3) 追蹤眼睛的位置 透過-個立體攝影機對用於執行眼睛位置辨識功能的計算模組提供一 對立體影像。在經職組運算後，模組會回傳每個眼睛相騎—個固定點 (例如《的中心點）的x、y、Az軸座標。這些座標可以（例如）透 過-個序列介面傳輸。為執行這個程序所_算可能由位於與顯示器像 素^在的同-基板⑽電·件（包括位於像素轉中㈣馳件） U i，來執行。 為了追縱-她看者的_，SLM面板上的全像蝙财以在X及/或 67 200916986 y軸方向上（也就是在面㈣平社）移位。根據所制的全像編碼方法 像編彻容在SLM上朝X 或y軸方向移位來執行-個側向方向的眼睛追縱。在進行則的全像編 瑪之前，計算模組會計算出全像影像資料相對於SLM在x或y轴 的偏移量。觀看者的眼睛的x、和 π 釉座松會作為一個輸入提供。 為了追縱-位觀看者的眼睛，SLM面板上的全像編碼可以在X及/或 y軸方向上（也就是在面板的平社）移位。追縱也可以透過使持續照亮 SLM的光源隨著觀看者位置的改變而同步移動來執行。不論發光的光源是 否會移動，或者-朗光線是縣辨_孔而㈣—致的光線照射的點 狀光源或直線光源中產生，只要通過這個開孔的光都會被視為一致的光 線。如果礙^議哪物_，㈣細址並且可 以配合觀看者的位置即時調整。 0.像差修正 在某些種類的全像顯示器中，像差修正是用來修正因負責執行傅立草 轉換的-個凹凸透鏡陣列中、或—個2D透鏡陣附的各個透鏡所造成的 像差。像纽應會峡傳播顺看者的方向與光軸之間的角度而異，並且 可以透過空間光調制㈣編碼來進行㈣H修正運算可以獨立於全像 計异之外錄執行’—朗產线和全像__止。_步驟之 後’總和全像影像和像錄正對射以—起進行調制。 68 200916986 像差正運异可以透過分析方式或利用檢視表（册）來執行。最終的 全像影料#録好只在有總和全像影像可供使㈣後魏複乘法來調 制。關於讎正的-麻__㈣犯巾麻。娜3中，像 差修正會利繼輪輪物㈣输縣_。不過，# 他的情況下’像差修正财能酬位於像素_所麵郎以外的位置、 但在與像素矩基板上㈣馳件來實施。 P·光斑修正 在某些種_全雜轉中，光哺正㈣來齡__顯示器的 不同區域之_光學-致性程度過柄造成的光斑。光斑效應可以透過空 間光調制_縣輯_修正。修正運討_立於全料算之外並 行執行，-直到產生總和全像影像的步驟為止。在這個步驟之後，總和全 像影像和光斑修正對應可以一起進行調制。 光斑修正運算可以透過分析方式或利用檢視表⑽）來執行。最終的 全像影料算録好只林總和全像雜可供使心紐職乘法來調 制。關於光斑修正的-個顯示設計的範例如圖33中所示。在圖33中，光 斑修正會·位於像素矩_在的㈣巾的電馳縣執行。不過，光斑 也可能湘位於像素矩陣所在的空間以相位置、但在與像素矩陣同 —基板上的電路組件來實施。 69 200916986 Q.全像顯示ϋ的數位版„理技術（應） 會接I:—個全像顯示器的内容龍可能有受職保護，也就是顯示器 會接收到麵.密_轉料。高減數軸雜護 個針對在2D顯示琴卜每# nDU ； ^ 清晰度多脾八 同標準。具備有瓣解碼功能的高 1板⑽Γ面（HDMI)接餅2D顯示輯子系統的印刷 ==上：_上的—個基本弱點是從顯示器電子系統到面板 輪通&quot;疋在解碼之後進行。所以它可以透過將面板的資料傳 輪电路組件的電路接通來截取解碼過的資料。 一井L的電路組件來執行的。在本發日⑽-個顯示設計的-個進 巾’解碼和全像影料算則是分散在像素轉巾的電路组 取財=散的崎來執行的。因此面板上並不會有任何-個位置可以截 則解碼。如果麵板上_區_不_碼密錄， 、 變得更加困難。因為面板上並沒有任何連接器可以Pipeline), like the various ships that execute the coloring engine, transfers the Wei_patented graphics card to the TFTs installed in the LCD panel. In addition, according to the above concept, the king of the slain of the slain can be a display that performs holographic calculation in the space of the pixel moment. Or, in the same system, according to the above concept, the whole turn of the case can also be - the accounting is in the pixel to the face of the face, but can be _ exist on the same substrate as the pixel matrix on the substrate Calculated display. i K. 2D □ 3D conversion In the case of a 2D □ 3D conversion, the first image and the second image from the stereo image are transmitted to the display device, all or at least one Part of the hologram is performed in the space where the pixel is located, or on the substrate where the pixel is located. 2D □ 3D conversion calculation can be performed on the pixel where the pixel is turned, or on the substrate where the pixel is located On the 'or can _ to produce money regulations corresponding __ Lin should be transmitted to the display (four) road components, or can be carried out on the rest of the equipment located in the «Cong technology is clear. The mapping of the two transmissions is the difference between the two stereo images, and the image will usually only require less data than the complete image. If it is a 3D stereoscopic display, the first The image itself can be calculated by the difference between the current image and the image from the early time. Similarly, the second image may also be between the current image and the image from the early time. The difference is 61. The display device may then use a known calculation program used in this technical field to convert between 2D and 3D y images, and calculate a corresponding data based on the received data. The buffer corresponds to the two-dimensional (10) image of the data. If it is a color image, △ 曰 and there should be three rib knife color images with the corresponding data in the two original colors and corresponding buffers. The data corresponding to the 2D image and buffer may then be processed by the device used to display the holographic image. This device will encode the holographic image on its SLM. For a full effective transmission bandwidth, The data transmitted in this system may be compressed by the known ship's private sequence' and the corresponding disassembly procedure is performed on the display splicing. Circuit components in which 2D □ 3D conversion is performed may have the right Accessing a repository containing a known set of 3D shapes in which it may attempt to match the 3D data it calculates; or it may have access to it - containing a known set of shapes The database, in this data oblique it may try to match the input 2 〇 image data. If there is a good match between the known shape, it will speed up the calculation process, because the rib or 3D image may be followed by a Relatively known shapes are represented. The 3D shape database may be transferred to groups such as sports stars (like top tennis stars or football stars) such as face P or body 雑' and joy scales (like Any shape or partial shape of a knowing course or a well-known football field. For example, an image of a person's face may be accessed by a display device, plus a change in facial expression (for example, it may be Smile or enemy eyebrows, plus some changes in the length of the hair (because the length of the hair can be long or short after obtaining the stored information). If the information that the display device has access to is very obvious, there is a difference of 62, 2009,986, which makes it interesting to mark the "radiation" of the "face", for example, - the individual's stagnation has changed and the materialization has been stored. - The time of the segment report can be updated by the display device. If the circuit component used for the calculation encounters a shape such as or imaginary _ in its ownership access to the job, the shape can be well matched, then these new shapes can be added to the record order. Adding a phantom loss can also be performed using a program known in the art to perform such conversions. Then: The 3D image data (buffer corresponding data and color corresponding data) is transmitted to the display to perform holographic calculation and display. The sub-U 3D conversion may be applied to the circuit that is to be displayed on a holographic display in which the omni-directional image is calculated in the space in which the pixel matrix is located, and the order is ordered. 'Or at least some of the holographic calculations will be performed on the electrical components placed in the space in which the pixel matrix is located, or at other locations on the substrate on which the pixel is located. L. Video Conversation (3D SkypeTM) According to the E3660065 Ship Alliance Community Trademark (EU c Kiss Kiss (4)) application, we know that SkypeTM can be given a (four) system secret to provide secret voice point to point (Qing) secret And age sharing, and instant messaging services; that is, providing communication », _ sharing, and (4) messaging services through a computer network. Apply for the 63 200916986 case under Section Ε 108 Ε Ε Ε Ε Ε Ε Ε Ε Ε Ε Ε M M M M M M M M M M M 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们 我们f material transmission, and instant messaging service computer software and hardware design; miscellaneous people build and rely on _; on the side of the global computer network computer servo red control others' installation and transfer software; provide online temporary Use, non-downloadable computer software to allow users to use the v〇Ip communication service; provide online software for others to download to allow users to use the v〇IP communication service. According to UK Trade Mark 2358090, we know that SkypeTM can provide Internet access, people π and fast secrets; telecommunications and telecommunications services; Internet Protocol ("IP") services; Internet voice Association ("VoIP") service; e-mail and Internet communication services; telecommunications services via third parties; Internet Protocol ("Ip") numeric telephone numbers and numeric telephone numbers for "IP" mapping systems and Database; domain and domain database system; rental computer data inventory time (provided by the Internet service provider). Any of the above services can be implemented in conjunction with a circuit component that can be implemented in a space in which the pixel matrix is located to perform all holographic calculations, or at least can be placed in a space in which the pixel matrix is located - part of the total A computing-like hologram display provides, in addition to the VOIP service provided by SkypeTM, an Internet Voice and holographic image protocol (VHIGIP) service. In this case, the above-described procedure can be green by the TFTs located in the liquid crystal panel. In addition, any of the above services can also be provided in conjunction with a reading age H that does not execute the full size of the (4) towel in the pixel matrix. In addition to the coffee, the provided V0IP County can also provide an Internet. Voice and hologram agreement (leg hidden) service. In addition, any of the above services can also be combined—a circuit component that does not perform holographic calculation in the space of the pixel matrix 64 200916986, but will be located on the same substrate as the pixel matrix. Calculated full-sided display to provide, in addition to coffee? In addition to the provided gift service, you can also provide an Internet voice and holographic image (side ιρ) service. In addition, the same 'Shanglan' service can also be combined with any holographic display, in addition to yp t, νοίΡ service, can also provide an Internet voice and holographic video protocol (vhioip) service. In addition, the above-mentioned service can also be combined with a holographic display that does not perform holographic calculation in the space where the pixel matrix is located. In addition to the provided gift service, it can also provide an internet voice and all Like the image protocol (v_(8) service. In the above (4) service, V_P can be provided in the form of (10) Internet voice holographic image protocol (VVHI0IP). Legs or simple ιρ can be provided in an instant or near-instant manner, And these Internet Nana can allow instant or near-instant holographic video communication between two people who are all in the image. M. Coding compensation in traditional photography, exposure fiber is fine - for possible Presenting - not awkward = like the side of the material 'to the _ gradually calculating the sister or the use of the transfer to provide compensation for the touch ^ butterfly Jing Qier (four) face (four) bribe, crossing 2 treatment 'fibre W. · _ towel The change of the speed of the flap 4 film, or other time to apply exposure compensation. In photography, 65 200916986 many cameras will include such a function so that the user can automatically adjust the calculated exposure value. Is divided into multiple segments of positive compensation (increased exposure) or negative compensation (reduced exposure) 'usually each segment will increase or decrease 1/3 or 1/2 grid aperture (f) setting, usually up to the increase (decrease) Two or three adjustments. Optically, from the point of view of the effective focal length of the lens, the f-number of an optical system represents the pupil (fine) opening and diameter. In the case of miscellaneous, the f-value is usually dispersed in multiples. The position of the light is called 'f' setting. Each "set" is marked with a corresponding f-number, and each-grid represents the light intensity equivalent to the front-lattice-half. This is quite unique. The square root magnification reduces the diameter of the pupil and aperture openings, thus halving the area of the pupil (aperture). Exposure compensation is used when the user knows that the camera automatically calculates the exposure value and will cause an undesirable exposure. - (4) bright _ sub-county scenes often create a lack of secret light, and: a dark _ sub-based scene often causes overexposure. - Under an experienced situation, this will happen, and know How much to apply After paying for the monthly payment, you will receive a photo with Yuan Wu exposure. ^ Any work described can be _ _ can be on the same side of the pixel - the first 3 Γ on the substrate, or at least on the side where the pixel is turned - Execution on the silk - part of the ten-rrr king image does not test the L s σ /, any of the above functions can also be combined with a pixel space on the same substrate on the silly matrix - Α 胄Wang Jing, or at least can be provided in the holographic display of the holographic image of the pixel matrix. In addition, any of the above functions 66 200916986 can also be provided in conjunction with any full-image age. The step is applied to the hologram image to provide an image that can be seen more clearly, that is, the image that the observer would normally consider to be correct, without the problem of overexposure. Ν. Eye tracking The holographic device may have an eye tracking function for one or more viewers. This is especially true for the viewing of each eye, which is especially a few millimeters. Xian preference system - the frequency of the material _ ^ tracking steps to the user's eyes: 1) to limit the search range by detecting the user's face 2) to limit the tracking range by detecting the eye 3) tracking the position of the eye through - A stereo camera provides a pair of stereo images to a computing module for performing an eye position recognition function. After the operation group operation, the module will return the x, y, and Az axis coordinates of each eye riding a fixed point (for example, the center point). These coordinates can be transmitted, for example, through a serial interface. In order to execute this program, it may be performed by the same-substrate (10) device (including the pixel-to-pixel (four)) U i located at the same time as the display pixel. In order to trace - her viewer's _, the hologram on the SLM panel is shifted in the y-axis direction of X and / or 67 200916986 (that is, in the face (four) Pingshe). According to the holographic coding method, the image tracking is performed on the SLM in the direction of the X or y axis to perform eye tracking in a lateral direction. The calculation module calculates the offset of the holographic image data relative to the SLM on the x or y axis prior to performing the holographic programming. The x, and π glaze pins of the viewer's eyes are provided as an input. In order to track the eyes of the viewer, the holographic code on the SLM panel can be shifted in the X and/or y-axis directions (i.e., in the panel). The tracking can also be performed by causing the light source that continuously illuminates the SLM to move synchronously as the position of the viewer changes. Whether the light source of the illuminating light will move, or the ray light is generated by the light source or the linear light source illuminated by the light, as long as the light passing through the opening is regarded as a uniform light. If it is a matter of _, (4) the address and can be adjusted immediately with the position of the viewer. 0. Aberration correction In some types of hologram displays, the aberration correction is used to correct the lens attached to a lenticular lens array or a 2D lens array. Aberration. Like the angle between the direction of the wave and the optical axis, the New York-Gap gorge can be transmitted through the spatial light modulation (4) coding. (4) The H correction operation can be performed independently of the holographic calculation. And the whole picture __ stop. After the _step, the sum hologram and the video are being modulated in the same way. 68 200916986 Aberrations can be performed by analysis or by using a view table. The final hologram #record is only available if there is a total hologram image available for (4) Wei Wei multiplication. About Yongzheng - hemp __ (four) commits linen. In Na 3, the aberration correction will benefit the wheel (4) to lose the county _. However, # in his case, the aberration correction property is located at a position other than the pixel _, but it is implemented on the pixel matrix (4). P. Spot Correction In some kinds of _ full miscellaneous rotation, the light is fed (4) to the age of __ different areas of the display _ optical-induced degree caused by the handle caused by the handle. The spot effect can be corrected by spatial light modulation. The correction is to be carried out in parallel with the entire calculation, until the step of generating a total holographic image. After this step, the sum hologram and spot correction can be modulated together. The spot correction operation can be performed by analysis or by using the view (10). The final hologram is recorded, and only the total sum of the forests and the whole image can be used to make the new multiplication. An example of a display design for spot correction is shown in FIG. In Fig. 33, the spot correction is performed at the Pitch County where the pixel moment _ is in the (four) towel. However, the spot may also be implemented in the space where the pixel matrix is located at the phase position, but in the same circuit block as the pixel matrix. 69 200916986 Q. The full-image display digital version of the digital technology (should be connected) I: - the content of a full-image display dragon may be protected, that is, the display will receive the surface. The number of axes is designed for the 2D display of the piano. Each # nDU; ^ The clarity of the spleen and the same standard. The high-plate (10) Γ (HDMI) with the petal decoding function is printed on the 2D display subsystem == The basic weakness of :_ is from the display electronics system to the panel wheel pass &quot;疋 after decoding. So it can intercept the decoded data by turning on the circuit of the data transfer circuit component of the panel. The circuit components of L are executed. In this (10)-one display design, the 'deletion' decoding and the holographic image calculation are performed by the circuit group of the pixel sub-division. Therefore, there will not be any position on the panel that can be truncated and decoded. If the _ area_ not _ code is recorded on the panel, it becomes more difficult because there is no connector on the panel.

板的解碼的資料，那些想要規避DRM的人必須必須知道面 ♦ t /1連接廣泛分散在整個顯示器上各的部位的多個TFT :^^存取經過解碼的資料。這對於改善腦的保護確實有非常卓著 會利tr侧示設計的—個進—步的範例是，解碼和全像影像計算 於像素矩陣所在的基板上的電路組件（包括位於像素矩陣以外的 70 200916986 位置上的電路組件）來執行。本發明的—個顯示設計的—個進—步的範例 是’解瑪和咖㈣會猶罐嶋__板上的電路 組件（包括位於像麵_外的位置上的電路組件）來執行。 解碼 R. 2D顯示器的數位版權管理技術（醜） 供應給-個2D顯示器的内容資料可能有受麵保護，也就是顯示器 會接收到有編碼域_容倾。高頻寬數簡容賴齡（耐）是一 個針對在2D顯示器上實施麵的共同標準。具備有騰p解碼功能的高 清晰度多媒體介面⑽Π接收器通常位於2d顯示器的電子系統的印刷 電路板（P⑻上。傳峨終悔輪是彳獅晴_面板 的影«料傳輸通常是在解碼之後進行。所以它可赠過將面板的資料傳 輸％路組件的電路接通來截取解碼過的資料。 在本發明的-個顯示設計的—個範例中，解碼是利用分散在整個則 面板上的電馳件以-個分散的崎來執行的。面板上並不會有任何 -個位置可喊_有峨着㈣。峨面板上的權位採用不 同的解碼紐，則解瑪絲的破解將會變得更加雜。因為面板上並沒有 任何連獅可以用來從面板截取經過解碼的資料，那些想要規避麵的人 必須必須知道面板的魏圖並且傾連接歧分散祕麵示紅各的部 位的多個TFT電晶體才能存取經過解物料。這對於改善麵的保護 確實有非常卓著的貢獻。 71 200916986 在本發明的-_ _+的，—步的_巾，I—㈣顯㈣ 置會用來透雜_|7，謝（.秘_料或者在 像素矩陣以外）的電路組件來執行解碼計算。這樣的電路組件比位於顯示 器的PCB上的電路組件更難以存取。這也有助於改善醜的保護。 S•在與顯示器以實體線路連接的硬體中執行軟體應用程式 原則上，電腦軟體的許多部份也可以利用電腦硬體來獨立執行。在本 Γ月的—個顯示設計的—個範射，—個可以彻軟體來執行的應用程式 經改為·分散在—個SLM面板的整個基板上的電路組件在硬體中執 艰__咐，樹驗_素矩陣所在 口 2板上但在像素矩陣以外的位置。SLM面板可以是用於全像顯示 或者用於2D顯示器的SLM面板。 T. 採用多個微稜鏡的可變光束轉向 看者❿追職看者顿看者崎触置的全雜㈣，投向觀 加以=雜置的“可縣束細，，會獨光束進行可 轉向的微稜鏡陣列來執行。而這個“可加以控制的轉向，，則是 22的。魏會透過一個位置_和追«統來執行。稜鏡的性質會 在—個或兩個維度上進行轉向的梅進行控制。兩個維度的 72 200916986 光線轉向可以利用縱列的兩個微稜鏡陣列來進行，例如，將其中一個陣列 中的稜鏡的縱軸設定在與另一個陣列中的稜鏡的縱軸呈一個有效角度的方 向’例如大約90。。針對不同的應用，這樣的幾何配置考參見第 US4, 542’449號專利中的敘述（合併在本文件中做為參考）。圖顯示出 光線會根據稜鏡的性質來以一個較小或以一個較大的角度進行偏導。這些 稜鏡可能是可以根據所施加的電荷來改變偏導角度的液態微稜鏡〔例如 採用電濕潤微稜鏡進行靈活的廣角光束轉向”文中所述，Heikenfeld等 著，Optics Express 14，pp. 6557-6563 (2006)，（合併在本文件中做為 參考）〕，或者是可以控制光束轉向的其他已知的稜鏡陣列。 如圖34中所可見的，平行的光線在通過SLM和稜鏡遮罩時會根據稜 鏡的性質而被偏導。這個程序的—個優點是，可以在光線通過稜鏡之前先 行減少各種光學效應（例如透鏡的像差）。這種方法適合用於將v〇w置於 觀看者或觀看者眼睛處的應用上。在另外的一個範例中，在稜鏡陣列之前 或之後加設一個聚焦的手段（例如一個傅立葉透鏡陣列）將可以協助將光 線匯集到vow上。 當-個觀察者改變他的位置時，稜鏡的偏導角度可以隨之進行調整（例 如透過調整施加在液態微稜辦壯的電壓）。這個偏導缝是可以連續改 變的。而這些稜鏡並不-定都有相_偏導肖度。另外，它也可以針對每 -個稜鏡侧進行控制’使每-個稜鏡可以有不關偏導驗（例如用來 進行Z軸追蹤），也就是使賴銳_的親可以大致匯_ 處， 73 200916986 靠近顯示器或更 因為丽與顯示器之間的距離將會因為觀看者移動到更 遠離顯示器的位置而改變。 稜鏡角度的計算在輯时齡將制相位崎慮。稜鏡角度 的計算可齡由位於SLM的基板上的運算電驗件來執行（轉物件㈣ 重建-樣），或者卿設置在稜鏡_基板上的運算電路組件來執行。而如 果SLM的基板也可以用來做為稜鏡陣列的基板，則並不須要有—侧立的 稜鏡陣列基板。 ^ 位準裝置與SLM之間需要有一個通訊介面：例如，這個介面可以 是一個序列介面。 如果用於計算稜鏡_偏導角度的運算電馳料是在稜鏡陣列的基 板上’而是在SLM㈤基板上，則兩個基板之間需要有—個資料連接線路， 使棱鏡陣列的電極可以利用計算的結果來進行控制。 除了用於控制稜鏡的計算之外，我們還必須施加一個“相位修正，，針 對因稜鏡陣列所造成的相位“跳動，，（或稱“相位不連續，，）的現象提供 補償。否則稜鏡陣列所呈現出來的效果會像是一個閃耀光栅—般，也就是 通過不同棱鏡的波前的各個部份會有不同的光程距離到達v〇w處，因此呈 現出來的效果會像是一個格柵一樣，而改變稜鏡角度會影響到不同的繞射 級上所分布的能量。這_位修正可以由SLM在它的全像影像編碼功能之 外執行。通過這兩個組件（也就是稜鏡陣列和SLM)的光線根據每一個組 件的功能會進行一項複乘法。經過修正的相位對應資料中包括有微稜鏡陣 列所需的相位修正：全像影像會以用來重建物件點的代表SLM像素格狀態 74 200916986 的值，包括相位修正的值，來進行編碼。 以上所述的功能也可以套用在全像影像是在一個投影式設備中產生的 情況’這裡所稱的投影牽涉到在稜鏡陣列上構成一個SLM影像，而所需的 3D景象的重建則發生在V0W的鈾方，據以形成—個相當於在這個技術領 域中已知的投影設備的設備。所需的計算和設備類似於以上所述，對熟悉 這類技術的人自然熟知而不必贅言。必須針對棱鏡陣列中的各個棱鏡的# 導角度、以及用來修正因此所造成的相位不連續的相位補償進行計算。棱 鏡陣列的相位補償可以在於稜鏡陣列上形成SLM日寺提供、或者以—個置於 稜鏡陣列附近的額外SLM來分開提供。為了可以進行投影，Μ可能是可 以透光的而稜鏡陣列則可以反射、或者Μ可能是可以反射的而棱鏡陣列 則可以透光’對熟悉這類技術的人自然熟知而不必資言。 液態微稜鏡如“採用電濕潤微稜鏡進行靈活的廣角光束轉向”文中所 述〔Helkenfeld 等著，0ptlcs Εχρ_ 14，卯 655? 6563 (細6)，（合 併在本文件中做為參考）〕。這個技術被稱為“電濕潤，，（❿伽讀也 或e-wet伽）。在這項技術中，由一種透明導電液體和另一種流體（例如 空幻之_編與-㈣_她纖觸成的接觸角是施 加在電極上的電壓差之於透明導電液體的_個函數。個別控制施加在兩個 各包覆有不議、峨㈣_輪娜成—峨崎素格的側 壁，母-侧鶴♦個電觸_—麵_蝴上的電壓， 了 乂用來控制&amp;個接觸肖’並由此控制當光束在穿越像素格時的轉向。利 用電濕潤稜鏡來達到光束轉向目的的其他配置對熟悉這類技術的人自然熟 75 200916986 第一製造程序概述 體顯示裝置的基本結構申，提供 在本發_-個顯示設計的薄膜半導 有-個顯示部件，射的電路組件佈置在顯示料的各雜素之間的空 =成者在基板上的其他位置，时執行與將諸畅魏置的顯示部件 關的各種料。顯示部件、以及在顯示部件範_或者在基板上的盆 ，整蝴输上。喊他用來驅動 .扣部件的電馳件則可能佈置麵示部件的週邊，但仍整合铜—基板 用來操作空間光調制器的TFT電路組件，以及諸如用來執行邏輯運算 、、他電路、、耕，則可能以_個如以下所述的方法佈置在—個基板上，這 ^法類似於第戰153，咖號專利中所述用來製做—個不同的袭置構 w的方法，US6’153,893號專利整體合併在本文件中作為參考。其他的方 十…員技術的人自然熟知而不必贅言。這個基板可能是-個大面積 的基板，且基板可能是—種合適的玻璃材f_。使財板，其經常 採用的處ί!程序傾向於偏好低溫處理，至少就Si裝置製造技術的標準而 •^而諸如在大約画T進行用來產生裝置的間絕緣層的獨高溫氧化 處理程序_與低溫處理程料相容，這做理程序型溫度大約 76 200916986 侍從350°C到700X的範圍。 像素電極和用來進行開/關的薄膜電晶體排列在顯示部件中的一個矩 陣内。用來構成電路元件的薄膜電晶體則佈置在顯示部件的各個像素之間 或者在基板上的其齡置，或者可以舰在整合铜—餘上的顯示器驅 動部件中。薄膜電晶體可岐下閘極式，包括有—個閘極、—個形成在間 極上的-個絕緣層上的多晶半導體層、以及一個包括有在多晶半導體層上 形成的-個源極和-個没極的高濃度雜質薄膜。用來進行開/關的Μ可 能有-個輕度摻雜騎極⑽）構造，射會在多晶半導歸和高濃度雜 質薄膜之間插入一個低濃度雜質薄膜。 在-個典型_示設計巾，齡料有—她財像素電極的上側部 份、-個包括有用來進行開/關的TFT的下側部份、以及可能有一個遽色 層、-個黑色遮罩層、和-個插入在上側和下側之間的平坦化層。若是這 種清況，黑色遮罩層會包含有一個金屬線層以便與源極和没極的高濃度雜 質層構成電氣連接。同時，像素電極也會透過金屬線層與汲極的高濃度雜 質薄膜構成電氣連接。此外，如果使用有—個包含有三原色並以一個時序 多工模式點亮的背光燈，則濾色層可以省略。 個具有以上所述的構造的顯示裝置可以糊下列低溫製程來製造。 百先’在玻璃基板上形成_。接τ來，在_上的—個絕緣薄臈上形成 —個半導體祕，然後透過雷射退火將半導體_轉魏—個多晶層。然 後選擇性的只在包括於像素開/關中的多晶層上形成-個低濃度雜質層， 例如透過_-個遮罩層。除此之外，也在低濃度雜質薄膜上形成—細 77 200916986 於源極和聰的高濃，並觀構成-個擁朗疊LDD構造、用來 進行開/關的TFT。在此同時，用於電路元件的财則會透過直接在包 触電路組件部份中（像是用於影像顯示計算、或用於週邊驅動部份）的 夕晶層上形成—烟於雜和馳㈣濃度雜㈣來構成。最好，可以在 =在電馳件部射的高濃度雜制上選擇性的執行雷射退火以便降低 多晶半導體層的電阻。 虽在-個_基板上形成_後’會低溫下在祕上的—侧極絕緣 薄膜上形成-辨導體薄膜。這辨導體軸接著再透過·退火轉變成 一個多晶層。邮，™_細啦《-«謂。所使用 的雷射典型上财-她賴波長使雷射_射可以在&amp;中被強力吸 收：一個鮮是_準分子雷射，但其他也坡已知的。由於它是-個下 間極式TFT’魏的結顧_物自雜請刷基板中的鈉） 所造成的不良影響。在裝置區域中使用多晶半導體層可以讓我們製造更小 的TFT。在鎌像素^關的抓中，咖的構造可以使韻電流保持在 低的Jc準彡ϋ属電流太高，則可能會在—麵示裝置中產生致命的 缺失。相反的’在構成電路元件的聊中，Ν通道型TFT和ρ通道型 TFT可以利用低狐處理程序透過在多晶半導體層上重疊一個高濃度雜質層 來同時形成。可以對構成電路耕的TFT執行額外的雷射退火來提高這些 TFT的速度。也可以扭用另外的一個構造，這個構造包括有一個渡色層、 一個黑色遮罩層、和—個平坦化層，以幫助達到更高的像素密度和更高的 開口率。 78 200916986 不限於TFT的構造 ’它同時也可 可以利用這個製造方法製成的構造並 以適用於於任何已知的構造。 第二製造程序概述 在本發明的一個顯示設計的薄膜半導體顯示裝置的基本結構中，提供 有-個顯示部件，其中的電路組件佈置在顯示部件的各個像素之間心 二，同—基板上的其他位置，用來執行與將資料顯示在I置的顯示: 有關的各種計算。顯示部件、以及用來執行計算的電路紐件，整合佈 Μ基板上。而其他絲鶴顯示部件的電路組制可能佈置在顯示料 的週邊’但仍整合在同—基板上。 用來操作空間光調制器的TFT電路組件’以及諸如用來執行邏輯運算 的其他電路組件’則可能以—個如以下所述的方法佈置在—個基板上，這 ^方法類似於第US6，刚，667號專利中所述用來製做_個不同的裝置構 造的方法；US6,140, 667號專利整體合併在本文件付為參考。其他的方 法對熟悉這類技術的人自麵知Μ必贅言。可以這健造程序製成 的石夕的種_為“連續晶_”，而㈣電學雛在某些方面（或者說在 泎多方面）可能類似於單晶矽的電學特性。For the decoding of the board, those who want to circumvent DRM must know the face ♦ t / 1 to connect multiple TFTs that are widely dispersed throughout the display: ^^ access to the decoded data. This is a very good example of improving the protection of the brain. A further example is that the decoding and holographic images are calculated on the substrate on which the pixel matrix is located (including 70 outside the pixel matrix). 200916986 The circuit components on the location) to execute. An example of an advancement of the display design of the present invention is that the 'sma and the coffee (four) will be executed on the circuit components on the board (including the circuit components located outside the image plane). Decoding R. 2D display digital rights management technology (ugly) The content data supplied to a 2D display may be protected by face protection, that is, the display will receive a coded domain. The high-frequency wide-numbered version is a common standard for implementing faces on 2D displays. The high-definition multimedia interface (10) with the Tengp decoding function is usually located on the printed circuit board of the electronic system of the 2d display (P(8). The final repentance wheel is the shadow of the lion's _ panel. The material transmission is usually decoded. After that, it can be used to intercept the circuit of the data transmission % component of the panel to intercept the decoded data. In the example of the display design of the present invention, the decoding is spread over the entire panel. The motorized parts are executed in a discrete way. There is no position on the panel that can be shouted _ there is a squat (four). The weights on the panel use different decoding buttons, then the solution will be solved. It will become more complicated. Because there aren't any lions on the panel that can be used to intercept decoded data from the panel, those who want to evade the surface must know the panel's Weitu and pour the distracting secrets. A plurality of TFT transistors in the part can access the material to be decomposed. This has a very significant contribution to the improvement of the surface protection. 71 200916986 In the invention of the invention -_ _+, - step _ towel, I - (four) display (four) The circuit components used to perform the decoding calculations are used to perform the decoding calculations. Such circuit components are more difficult to access than the circuit components located on the PCB of the display. Helps to improve ugly protection. S• Execute software applications in hardware connected to the display with physical lines. In principle, many parts of the computer software can also be executed independently using computer hardware. Displaying the design of a fan, an application that can be executed in software is changed to · The circuit components scattered on the entire substrate of the SLM panel are hard in the hardware __咐, the tree test _ prime matrix The position on the board 2 but outside the pixel matrix. The SLM panel can be an SLM panel for holographic display or for 2D displays. T. Variable beam steering with multiple micro-turns to turn the viewer The whole miscellaneous (four) that is touched by the watcher, and the view to the miscellaneous "may be fine, the single beam will be executed by the steerable micro-array array. And this "controllable steering," It is 22. Wei Huitong A position _ and chase system are executed. The nature of 稜鏡 is controlled by the steer in one or two dimensions. The two dimensions of 72 200916986 ray turn can use the two micro-array arrays of the column. For example, the longitudinal axes of the turns in one of the arrays are set at a direction that is at an effective angle to the longitudinal axis of the turns in the other array, for example about 90. For such applications, such geometric configurations are tested. See the description in U.S. Patent No. 4,542,449, the disclosure of which is incorporated herein by reference in its entirety in its entirety in the the the the the the the the the the the稜鏡 It may be a liquid micro-turn that can change the deflection angle according to the applied charge (for example, flexible wide-angle beam steering using electrowetting micro-twisting), Heikenfeld et al., Optics Express 14, pp. 6557 -6563 (2006), (combined in this document as a reference), or other known array of turns that can control beam steering. As can be seen in Figure 34, the parallel rays are deflected according to the nature of the prism as they pass through the SLM and the 稜鏡 mask. One advantage of this program is that it can reduce various optical effects (such as lens aberrations) before the light passes through the 稜鏡. This method is suitable for placing v〇w on an application at the viewer or viewer's eyes. In another example, the addition of a focusing means (e.g., a Fourier lens array) before or after the 稜鏡 array will assist in concentrating the light onto the vow. When an observer changes his position, the angle of deflection of the helium can be adjusted accordingly (for example, by adjusting the voltage applied to the liquid micro-edge). This partial guide can be continuously changed. And these 稜鏡 do not have a phase 偏 bias. In addition, it can also be controlled for each side of the ' ' so that each 稜鏡 can have a non-biased guide (for example, for Z-axis tracking), that is, the relatives of Lai Rui _ can be roughly _ , 73 200916986 Close to the display or because the distance between the MN and the display will change as the viewer moves to a position further away from the display. The calculation of the angle of the 将 will be phased in the age of the series. The calculation of the 稜鏡 angle can be performed by an arithmetical test piece located on the substrate of the SLM (transfer object (4) reconstruction-like), or by an arithmetic circuit component disposed on the 稜鏡_substrate. If the SLM substrate can also be used as a substrate for a tantalum array, then there is no need for a side-by-side tantalum array substrate. ^ There is a communication interface between the level device and the SLM: for example, this interface can be a serial interface. If the calculation for calculating the 稜鏡_bias angle is on the substrate of the 稜鏡 array, but on the SLM (five) substrate, a data connection line is required between the two substrates to make the electrode of the prism array The result of the calculation can be used to control. In addition to the calculations used to control 稜鏡, we must also apply a “phase correction” to compensate for the phenomenon of phase “jumping, (or “phase discontinuity,”) caused by the 稜鏡 array. The effect of the 稜鏡 array will be like a blazed grating, that is, each part of the wavefront of different prisms will have different optical path distances to reach v〇w, so the effect will be like A grid is the same, and changing the angle of the 会 affects the energy distributed across the different diffraction levels. This _ bit correction can be performed by the SLM outside its holographic image encoding function. Through these two components (also The ray of the 稜鏡 array and the SLM) performs a complex multiplication according to the function of each component. The corrected phase correspondence data includes the phase correction required for the micro 稜鏡 array: the hologram image is used to reconstruct the object The point represents the value of the SLM pixel state 74 200916986, including the value of the phase correction, for encoding. The above described functions can also be applied to the holographic image in a projection The situation generated in the preparation 'The projection referred to here involves the formation of an SLM image on the 稜鏡 array, and the reconstruction of the required 3D scene occurs at the uranium side of V0W, which is equivalent to forming in this technical field. Known projection device devices. The required calculations and devices are similar to those described above, and are well known to those skilled in the art and need not be rumored. It is necessary to target the angles of the individual prisms in the prism array and to correct them. Therefore, the phase discontinuity phase compensation caused by the calculation is performed. The phase compensation of the prism array may be provided by forming SLM day temple on the 稜鏡 array or separately by an additional SLM placed near the 稜鏡 array. Projection, Μ may be light transmissive and 稜鏡 arrays may reflect, or Μ may be reflective and prism arrays may transmit light' is familiar to those familiar with such techniques and does not need to be eloquent. For example, "Flexible wide-angle beam steering with electrowetting micro-turns" is described in the article [Helkenfeld waits, 0ptlcs Εχρ_ 14, 卯 655? 6563 ( 6), (merged by reference in this document)]. This technique is called "electrowetting ,, (❿ gamma read or gamma or e-wet). In this technique, the contact angle between a transparent conductive liquid and another fluid (e.g., phantom-------------------------------------------------------- The individual control is applied to the two sides of the side wall, the side wall, the mother-side crane ♦ the electric contact _-face _ butterfly, and the 控制 is used to control &amp; a contact Shaw' and thus control the steering of the beam as it traverses the pixel. Other configurations that use electrowetting 稜鏡 to achieve beam steering are naturally familiar to those familiar with this type of technology 75 200916986 First Manufacturing Procedure Overview Body Display The basic structure of the device provides a thin film semi-conductor having a display component in the present invention, and the circuit component of the projection is arranged between the various materials of the display material; the other position on the substrate At the time, the various materials that are used to close the display parts of the display are executed. The display part, and the display part or the basin on the substrate, are all turned on. The driver is used to drive the buckle parts. It is possible to arrange the periphery of the component However, the TFT circuit assembly, which is still integrated with the copper-substrate for operating the spatial light modulator, and such as for performing logic operations, other circuits, and ploughing, may be arranged on a substrate in a manner as described below. This method is similar to the method described in the No. 153, the coffee number patent, which is used to make a different attack structure. The US Patent No. 6,153,893 is incorporated herein by reference in its entirety. The person skilled in the art is naturally familiar and does not have to be rumored. This substrate may be a large-area substrate, and the substrate may be a suitable glass material f_. The financial board, which is often used in the program, tends to prefer low temperature processing. At least as far as the standard of the Si device manufacturing technology is concerned, such as a separate high temperature oxidation process for producing an insulating layer of the device at about T, which is compatible with the low temperature process material, the process temperature is about 76. 200916986 The range of 350 ° C to 700 X. The pixel electrode and the thin film transistor for on/off are arranged in a matrix in the display part. The thin film transistor used to form the circuit element is arranged in Displaying the individual pixels of the component or on the substrate, or it can be embedded in the copper-over-the-top display driving component. The thin film transistor can be mounted on the gate, including a gate, a a polycrystalline semiconductor layer formed on an insulating layer on the interpole, and a high-concentration impurity film including a source formed on the polycrystalline semiconductor layer and a non-polarity for opening/closing Μ There may be a lightly doped rider (10) structure that inserts a low concentration impurity film between the polycrystalline semiconducting and high concentration impurity films. In a typical design towel, the age of the material has the upper part of the pixel electrode, the lower part of the TFT including the opening/closing, and possibly a enamel layer, a black The mask layer, and a planarization layer interposed between the upper side and the lower side. In this case, the black mask layer will contain a layer of metal wire to make electrical connections to the source and the high concentration of impurity layers. At the same time, the pixel electrode is also electrically connected to the high-concentration impurity film of the drain through the metal wire layer. In addition, if a backlight containing three primary colors and illuminated in a time-multiplexed mode is used, the color filter layer can be omitted. A display device having the above-described configuration can be fabricated by the following low temperature process.百先' formed on the glass substrate. Connected to τ, a semiconductor secret is formed on an insulating thin crucible on _, and then the semiconductor is transferred to a polycrystalline layer by laser annealing. Then, a low-concentration impurity layer is selectively formed only on the polycrystalline layer included in the pixel on/off, for example, through a mask layer. In addition, it is also formed on a low-concentration impurity film - fine 77 200916986 in the source and Cong's high concentration, and constitutes a TFT that is used to perform on/off. At the same time, the money for the circuit components is formed by directly forming the layer of the circuit component (such as for image display calculation or for the peripheral driving part). Chi (four) concentration of impurities (four) to constitute. Preferably, the laser annealing can be selectively performed at a high concentration miscellaneous shot of the portion to reduce the resistance of the polycrystalline semiconductor layer. A thin film of a conductor is formed on the side-side insulating film at a low temperature even after forming a _ on the substrate. This discriminates the conductor axis and then re-permeates and anneals into a polycrystalline layer. Post, TM_fine "-« said. The laser used is typically on the fortune - she relies on wavelengths to make the laser ray shots that can be strongly absorbed in &amp; a fresh _ excimer laser, but others are also known. Since it is a negative effect of the lower-pole TFT'Wei's contact with the sodium in the substrate. The use of a polycrystalline semiconductor layer in the device area allows us to make smaller TFTs. In the grasping of the pixel, the construction of the coffee can keep the rhythm current at a low Jc. The current is too high, which may cause a fatal defect in the device. Conversely, in the case of constituting circuit elements, the Ν channel type TFT and the ρ channel type TFT can be simultaneously formed by superposing a high concentration impurity layer on the polycrystalline semiconductor layer by a low fox processing program. Additional laser annealing can be performed on the TFTs that make up the circuit to increase the speed of these TFTs. It is also possible to twist another configuration that includes a color crossing layer, a black mask layer, and a flattening layer to help achieve higher pixel density and higher aperture ratio. 78 200916986 is not limited to the construction of the TFT. It can also be constructed by this manufacturing method and is applicable to any known configuration. Second Manufacturing Procedure Overview In the basic structure of a thin film semiconductor display device of a display design of the present invention, a display component is provided, wherein the circuit component is disposed between the respective pixels of the display component, on the same substrate Other locations are used to perform various calculations related to displaying the data on the I: display. Display components, as well as circuit components for performing calculations, are integrated on the substrate. The circuit assembly of other silk crane display components may be arranged at the periphery of the display material but still integrated on the same substrate. The TFT circuit assembly ' used to operate the spatial light modulator' and other circuit components such as those used to perform logic operations may be arranged on a substrate in a manner similar to that described in US6. A method for making a different device configuration as described in the '667 patent; the entire disclosure of U.S. Patent No. 6,140, 667 is hereby incorporated by reference. Other methods are self-evident to those who are familiar with such technologies. The species of Shi Xi, which can be made by this health-making procedure, is "continuous crystal", and (4) electric chicks may be similar to the electrical properties of single crystal germanium in some aspects (or in many respects).

El 1 1 &quot;、12、和13顯示出一個可以用來形成適合用於顯示器的（包括 使用於像素開^/關、顯示器驅動、和邏輯電路組件中的）連續晶粒（[〇) 夕的流裎的簡述。基板n〇1可能是一個大面積的基板，且基板可能是一 79 200916986 種合適的_、或石英材_類。如《貞㈣m贿射的幾何配 置中（像基板並不須要可以透光的—個反射式幾何配置），則可以使用不透 土板例士原生夕日日石夕或陶究基板。基板有一個絕緣表面。薄膜no?是 一個非結晶石夕薄膜，其中石夕的厚度在1〇 nm们5 nm之間，包括所形成 的任何氧化物。&amp;個薄膜可以透過健化學蒸氣沈積⑽或透過電聚 CVD處理程序來生成。 以下’我們將制結晶料製程，但料其他的製程在這健術領域 中都疋已知的。先形成—層鮮絕緣薄膜腦，其中的開口對應於基板上 所而的CG料位置。以—種包含有Ni的溶液作為用來使糊一個旋轉 塗佈製程（在這個製程中會形成促進劑薄膜層）所塗佈的非結晶&amp; 晶化的促進劑。其他的促進劑像是c〇、Fe、Sn、pb、pd、pt、Cu、或如或 類似的物胃也都可以使用。在遮罩絕緣細11Q3中的開。處，促進劑薄 膜1104會與非結晶Si薄膜11〇2接觸。非結晶Si薄膜11〇2接著可 、透；L·在500 C到700。0之間的溫度下進行4小時到12小時的退火， 在一個惰性的環境中’或者是在—個包含有氫氣或氧氣的環境中生成結晶。 如圖11B中所示，在區域11〇5和11〇6中的非結晶Si 1102會因為 Νι促進劑的催化而結晶。接著形成水平生長區域和，這個生 長大致上會遍及整個基板。只有這些水平生長區域，如11〇7和11〇8，會 用來做為在基板上形成的TFT裝置的作用層。在退火完成之後，會將遮罩 層U03從基板上移除。接著進行成形，如uc中所示圖。這會在整個基 板上形成島狀的半導體層11〇9、111〇、和ιιιι(這些是作用層）。ii〇g是 200916986 構成-個互補金屬氧化半導體⑽s)電路的一個n通道型m的一個 作用層’ mo是構成一個CM0S電路的—個p通道型術的一個作用 層，而UU則是構成一個像素矩陣電路的—個N通道型μ的一個作 用層。 當作用層、和而已經形成後，接著會形成包括有一個 ^石夕的膜的-個閘極絕緣薄膜1112。_絕緣薄膜⑽的厚度 &quot;J 250 nm 5 2步驟中細-些氧化。薄膜1112可以利用已知的氣相生長方法來生 請_輪N1錢__ 素屬物質的環境中進行。它會在觸x :在3有_ 5), β , ^ j iUUU L之間的溫度下進行〇 1 到6小時的加熱。一個例子是在95〇 丁 0.1 _)的則的環境中（或者更通常的3人度下，在含有3容積百分比 間） 次者更通㊉的疋介於〇. 5㈣到10 ν〇1%之 (嶋心娜戦㈣謝_濃度的氮氣 緩薄膜中的石夕的氧化。 册、断、a2、F2、Bn、NF3、⑽、％卜其他蝴屬物質像是 個除氣程序可以用來移除薄膜上的Νι促進、明物質也都可以使用。這 形成的負各# s 促進蓟。這個作用似乎是透過讓所 絕緣_ :續揮發到環境谢而由環境空氣吸收來發生的。間極 _、和川的厚度在這個氧化_中會有增大的趨勢。區域·、 能增進場致移動率以月立°曰減低奶的闕閉⑽）電流，並且 移動率以及其他明顯的好處。 200916986 在經過上述處理之後，接著會在—個含有氮氣的環境中、在9,的 溫度下進行1小時的熱處理，以改善閘極絕緣_ 1112的品質以及問極 絕緣薄膜m2與區域測、m〇、和1U1之間的銜接面的品質。 接著會形成-個含有0.2重量百分比（wt%)的&amp;的M薄膜，並 且會形成-個用來構成-個閘極的原型的電極型樣（後面將會說明）。這在 圖11並沒有顯示出來。適合用於這個目的的其他材料，如Ta、w，、或El 1 1 &quot;, 12, and 13 show a continuous die ([〇) eve that can be used to form displays suitable for use in displays (including pixel on/off, display drivers, and logic circuit components) A brief description of the rogue. The substrate n〇1 may be a large-area substrate, and the substrate may be a suitable type of _, or quartz material. For example, in the geometric configuration of the 四(四)m bribe (as the substrate does not need to be able to transmit light – a reflective geometric configuration), you can use the impermeable earth plate, the original day, or the ceramic substrate. The substrate has an insulating surface. The film no? is a non-crystalline stone film in which the thickness of the stone is between 5 nm and 1 nm, including any oxide formed. &amp; films can be generated by chemical vapor deposition (10) or by electropolymerization CVD process. The following 'we will make a crystallizing process, but other processes are known in this field of health. First, a layer of fresh insulating film brain is formed, wherein the opening corresponds to the position of the CG material on the substrate. A solution containing Ni is used as a non-crystalline & crystallization promoter for coating a paste in a spin coating process in which a promoter film layer is formed in this process. Other accelerators such as c〇, Fe, Sn, pb, pd, pt, Cu, or the like may be used. Open in the mask insulation 11Q3. At this point, the accelerator film 1104 is in contact with the amorphous Si film 11〇2. The amorphous Si film 11〇2 is then permeable; L· is annealed at a temperature between 500 C and 700. 0 for 4 hours to 12 hours, or in an inert environment. Crystals are formed in an oxygen environment. As shown in Fig. 11B, the amorphous Si 1102 in the regions 11〇5 and 11〇6 crystallizes due to the catalysis of the oxime promoter. A horizontal growth region is then formed and this growth will generally extend throughout the substrate. Only these horizontal growth regions, such as 11〇7 and 11〇8, are used as the active layer of the TFT device formed on the substrate. After the annealing is completed, the mask layer U03 is removed from the substrate. Forming is then carried out, as shown in the figure uc. This forms island-like semiconductor layers 11〇9, 111〇, and ιιιι (these are active layers) on the entire substrate. Ii〇g is 200916986. An active layer of an n-channel type m that constitutes a complementary metal oxide semiconductor (10) s) circuit is an active layer of a p-channel type of CM0S circuit, and UU constitutes a layer. An active layer of the N-channel type μ of the pixel matrix circuit. After the active layer, and has been formed, a gate insulating film 1112 including a film of a stone is formed. _Insulation film (10) thickness &quot; J 250 nm 5 2 step in the fine-oxidation. The film 1112 can be carried out by using a known vapor phase growth method in the environment of the material of the genus. It will take 〇 1 to 6 hours of heating at a temperature between 3 and _ 5), β , ^ j iUUU L . An example is in the environment of 95 0.1 0.1 _) (or more usually 3 people, between 3 volume percentages). The second one is more than 〇. 5 (four) to 10 ν 〇 1% (嶋心娜戦(四)谢_ The oxidation of Shi Xi in the concentration of nitrogen slow film. Book, broken, a2, F2, Bn, NF3, (10), % Bu other material like a degassing program can be used to move In addition to the 促进ι promoting on the film, the bright substance can also be used. This forms a negative # s to promote 蓟. This effect seems to occur by allowing the insulation _: to continue to be evaporated into the environment and absorbed by the ambient air. The thickness of _, and Sichuan will increase in this oxidation _. Area, can increase the field-induced mobility, reduce the milk's stagnation (10) current, and the mobility and other obvious benefits. 200916986 After the above treatment, heat treatment is carried out for 1 hour in a nitrogen-containing environment at a temperature of 9, to improve the quality of the gate insulation _ 1112 and the polarity of the insulating film m2 and the area, m The quality of the interface between 〇 and 1U1. Then, an M film containing 0.2% by weight (wt%) of &amp; is formed, and an electrode pattern of a prototype for forming a gate is formed (to be described later). This is not shown in Figure 11. Other materials suitable for this purpose, such as Ta, w, or

Si也都可以使用。透過對這個型樣的表面進行陽極氧化，會形成問極 1113、1114、和1115 ’以及陽極化薄膜1116、lm、和1118，如仙中 所示圖。在下-個步驟中，如圖11E中所示，以佩方式絲薄膜⑴2 (例如使用CHF3氣體），使薄膜1112只會留存在各個電_正下方 1119、1120、和1121的位置。這時會以—個耐银遮罩1122來遮覆預定 用於-個P通道型TFT的區域。接著透過（例如）植人或電漿沈積方法 來加入用於η型材料的雜質離子’如圖UE中的箭頭所示。接著形成n型 區域1123、1124、1125'和1126。在這個過程之後，可以將耐姓遮罩1122 移除，並在η型區域上加上-個耐#遮罩1127來遮覆（圖12Α)。接著可 以透過（例如）植人或電漿沈積方法來對ρ型區域1128 # 1129進行換 雜。ρ摻雜區域也就是LDD區域。然射崎遮覆η韻域的耐钱遮罩 1127去除。 接著透過一個回韻處理程序在側壁113〇、1131、和1132上形成氧化 夕薄膜a j固遮罩1133來遮覆ρ型區域，並力口入η型摻雜劑來提高 &gt;又有被氧化物纖遮覆的區域巾的η型摻_的濃度。將雜力及極區域 82 200916986 的薄膜電_整到小於· Ω的水準，最好是小於· Ω。接著在間極 下方形成一個原生的或本質上是原生的通道成形區域1137。接著形成構成 像素矩陣魏的—讎極區域1138、-個祕輯1139、低濃度雜質區 域Π40、和—個Ν通道型TFT的通道成形區域1141(圖12C)。在圖12D 中’耐敍遮| 1133已經移除並在N通道型TFT上形成—個耐兹遮罩 1142 °進—步加人P型雜質來提高P型摻雜劑的濃度。接著將耐韻遮罩 1142移除並透過熱處理（例如徐冷爐退火、雷射退火、或類似處理）來使 雜質離子活化 這可以減少除因熱處輯造成的植入損壞。 接著形成一層厚度在20 nm到5〇 nm之間的Ή薄膜⑽並執行 抓用燈‘火方法的熱處理。Si肖Ti薄膜細會發生化學反應而形成石夕 化鈦，_成石夕化物區域1148、⑽、和·，如圖13A巾所示。圖ΐ3β 顯示出島狀的型樣⑽、腺，1153，這些型樣的形成可以避免石夕化 物薄膜區域1148、1149、和·在後續的步驟中由於形成絲連接源極 /沒極區域和配線的接觸孔而被去除。 接著形成一層厚度在〇·3 μηι到i㈣之間的氧化石夕薄膜作為第一中 介層絕緣薄膜脳。接著形成接觸孔，並形成源極配線1155、ιΐ56、和 1157以及汲極配線脳和，如圖13β中所示。可以使用有機樹脂 來做為第-中介層絕緣薄膜脳。在圖13C中，會在基板上形成—個厚度 在〇· 5 μηι到3 範圍内的第二絕緣層讓。可以使用聚亞酿胺、丙稀Si can also be used. By anodizing the surface of this pattern, the electrodes 1113, 1114, and 1115' and the anodized films 1116, lm, and 1118 are formed, as shown in the figure. In the next step, as shown in Fig. 11E, the film 1112 (e.g., using CHF3 gas) is allowed to remain in the position of the respective electrodes 1119, 1120, and 1121. At this time, a silver-resistant mask 1122 is used to cover the area intended for the -P-channel type TFT. The impurity ions for the n-type material are then added by, for example, implantation or plasma deposition as shown by the arrows in the UE. N-type regions 1123, 1124, 1125' and 1126 are then formed. After this process, the resistance mask 1122 can be removed and a mask #1127 can be added over the n-type region (Fig. 12A). The p-type region 1128 # 1129 can then be modified by, for example, implanting or plasma deposition methods. The p-doped region is also the LDD region. However, the anti-money mask of the 崎 遮 遮 η η 1 1 1 1 1127 removed. Then, through a reverberation processing program, a oxidized thin film aj solid mask 1133 is formed on the sidewalls 113 〇, 1131, and 1132 to cover the p-type region, and the η-type dopant is added to improve the oxidization. The concentration of the n-type doping of the tissue covered by the fiber. The film of the hybrid and polar regions 82 200916986 is electrically integrated to a level less than Ω, preferably less than Ω. A native or essentially native channel forming region 1137 is then formed below the interpole. Next, a drain region 1138, a secret 1139, a low-concentration impurity region Π40, and a channel forming region 1141 of a germanium channel type TFT constituting the pixel matrix Wei are formed (Fig. 12C). In Fig. 12D, 'Resistance| 1133 has been removed and formed on the N-channel type TFT - 1142 ° step-wise addition of P-type impurities to increase the concentration of the P-type dopant. The rhyme mask 1142 is then removed and subjected to heat treatment (e.g., quench annealing, laser annealing, or the like) to activate the impurity ions, which reduces implant damage due to heat. A tantalum film (10) having a thickness between 20 nm and 5 〇 nm is then formed and a heat treatment of the igniting lamp ‘fire method is performed. The Si-Sha Ti film undergoes a chemical reaction to form a titanium-titanium, and the yttrium-forming compound regions 1148, (10), and · are shown in Fig. 13A. Figure 3β shows island-like patterns (10), glands, and 1153. These patterns are formed to avoid the lithotripe film regions 1148, 1149, and in the subsequent steps due to the formation of wire-connected source/no-polar regions and wiring. The contact hole is removed. Next, a layer of a oxidized oxide film having a thickness of between 〇·3 μηι and i (four) is formed as the first intermediate insulating film 脳. Contact holes are then formed, and source wirings 1155, ι 56, and 1157, and drain wirings are formed, as shown in Fig. 13β. An organic resin can be used as the first inter-layer insulating film. In Fig. 13C, a second insulating layer having a thickness in the range of 〇·5 μηι to 3 is formed on the substrate. Polyaramine, propylene can be used

酸樹脂、魏胺、聚魏胺醯胺、或類崎料來做為械樹蘭膜。接著 在薄膜絕騎謂上形成-個黑色轉⑽。織形成—個厚度在U 83 200916986 μιη到0.3μιη範圍内的第三絕緣中介層薄膜1162,像是氧化石夕、氮化石夕、 氮氧化石夕、或-個有機樹脂薄膜、或是這些材料的層積薄膜。在薄膜謂 和薄膜1162上形成各個接觸孔’並形成一個厚度為12〇測的像素電極 1163。在-個黑色遮罩麗與像素電極聰重疊的區域形成一個辅助 電容1164，如圖13C中所示。 將整個基板在含氫氣的環境中在35(η：的溫度下加熱丨到2小 •時，這可以對懸浮鍵提供補償，尤其是在各個薄膜的作用層中。在這些步 驟之後’可以在同-基板上（例如在相鄰的位置上）形成圖既左側所示 的CMOS電路以及圖13C右側所示的像素矩陣電路。 可以透過_製造方法製成轉造並稀於τπ構造，而可以套用於 任何已知的構造，包括下閘極式TFT。 、 第三製造程序概述 在本發明的一個顯示設計的薄膜半導體顯示裝置的基本結構中，提供 有一個顯示部件，其中的電路組件佈置在顯示部件的各個像素之間的* 間，或在同—基板上的其他健，时執行與將資料置的顯示^ 件上有關的各種計异。顯示部件、以及絲執行計算的電路組件，整 的週邊，但健細爾蝴示部件 用來操作空間光調制器的TFT電路組件，以及諸 。、 的其他電路組件’則可能以—個如以下所述的方法佈 ^^ 個方法類餘請，则7細巾糊嫩,2置: 84 200916986 造的方法’· ·，759,677號專利整體合併在本文件中作為參考。其他的方 法對熟悉這類技術的人自麵知而不必贅言。可以_這種製造程序製成 的半導體麵是多晶雜，㈣的電學雜在某些方面（或者在許多方面 都）可能類似於（或甚至超過）單晶_電學特性。 這個製造程序可以使電路組件在同一個基板上形成。可以使用多晶石夕 做為作用層來產生許多可以用來控制顯示器的各個像素的聊。所產生的 其他TFT則可以具有像是_驅動電路、源_動電路、和訊號處理電路 的功能，在這些TFT中用來做為作用層的是石夕錯，以便提升高速運算的能 力。Ge會加入到需要具有高速運算能力的電路組件的部份，而多晶Si則 會用在需要具有低關⑽）電流特㈣電路部份。 在這個程序中會製造—她含有—個像素轉電路、収—個驅動電 路（在這鋪财這是-個GMQS電路）的絲_絲錢置，而這些 全部都在同-個基板上的—個絕緣表面上形成。這個處理程序如圖6中所 示〇 如圖6A中所示，準備—個玻璃基板6()1，然後在其上形成一層氧化石夕 6〇2。利用電衆CVD方法形成一個厚度為3Q nm的非結晶赠膜咖。利 用成形法在非結晶Si薄膜603上形成一個耐#遮罩6〇4。形成這個耐敍 遮罩是為了遮覆那些將用來形成用於像素矩陣電路的抓組群的區域。而 將用來形成高速電路的區域則不加以遮覆。如圖6β中所示，^會利用諸 士離子植人、電漿摻雜，，、或“雷射摻雜，，的技術來加入七的加 入疋為了改變非結晶si馳的组成成份以便產生—個均勻的sia薄 85 200916986 、… 1)的組成成份。如果採用離子植入技術，則要加入Ge的 薄膜605將會有遭受植入損壞的風險。因為πυ膜咖是處於一個 非結晶的狀態下。 由於在Ge中的晶格擴散所需的活化能量比在Si中為低，且Ge和 Si在一合金相位圖巾在低於雜以下的溫度下會成為彼此的—個固溶 體因此&amp;的存在可以用來加速Sh-xGex薄膜的結晶（相較於只有純Si 薄膜的結晶）。在IS ViEFi Γ J- 、方面，Ge在Si的結晶上（像是在雷射誘引結晶中） 可以被視為是一種觸媒半導體。 在圖6C中’耐蚀層6〇3已經被移除並在整個表面上加上一層含Ni 層606如第US5, 643, 826號專利中所述；（JS5, 643, 826號專利整體合併 在本文件中作為參考。Μ是用來做為一種觸媒材料以便力口速&amp;咬 Sll_xGeX薄膜的結晶。但除Ni以外的元素，像是Co、Fe、Cu、Pd、Pt、Au、 或In也可以用於這個目的上。如圖6])中所示，&amp;和％如薄膜的結 晶可赠過徐冷爐社來達成，在·C的溫度下社約M、時。這會 :形成一個多晶Sil.xGex區域6〇7和-個多晶石夕區域608。這個熱處理也可 以利用其他方法純行’例如像是雷射退火或燈照退火。 如圖6E中所不，多晶Sii xGex區域6〇7會在作用層咖巾形成。多 日日Si區域608會在作用層61〇中形成。其中作用層哪是用來做為將 會在後面構成-個驅動電路和訊號處理電路的TFT的作闕；而作用層 610則是用來做為將會在後面構成-個像素矩陣電路的TFT的作用層。 接著利用第US5, 648, 277號專射所述的—個處理程序來形成一個 86 200916986 源極區域、-做極區域、和—她度摻雜的祕（ldd)區域第 US5，_如號專利整體合併在本文件中做為參考。以下，我們將摘述這 個處理程序。首先’利用-個含有2 wt% # Sc力ai薄膜來形成—個稍 後將用來個的島狀難。接下來，對這個島狀雖執行陽極氧 化以便在島狀舰輸上形成—购_極氧蝴膜。接著改變溶 液來進-步執仃陽極氧化以形成—姆致的陽極氧化物薄膜環繞這些島狀 型樣。在〜個方式形成多孔性陽極氧化物薄膜和緊致陽極氧化物薄膜之 後，利用一個乾财法來侧閘極介電薄膜。在完成閘極介電薄膜祕刻 之後’將多錄陽極氧化物_除，藉此來獲_a中所示的狀能。 如圖7A中所示，、712、和？13是由氧化石夕薄膜所構成的間極絕 緣細’ 714、715，716 B m ”她物閉極，而 717 718彳719則疋用來保護閘極的緊致陽極氧化物薄膜。如圖7B中 所示，將咖成-個P通道型m的部位以―個遮請遮覆。兑餘 的部位則將進行η型離子植入以便提供η型導性。如第US5,648,277號 專利中所述，將會姻兩個不_加速電壓來使馳人轉子在整個深产 上的濃度有更均勻的分布。 在圖7B中，這個處理程序會形成时構成—個驅動電路的—個打通 道型τπ的-個沒極區域721、—個源極區域概、—個咖區域恐、 和一個通道區域724。同時也會形成聽構成—轉素矩陣電路的一個Ν 通道塑TFT的-個沒極區域726、一個源極區域概、—個⑽區域 727、和一個通道區域728。 87 200916986 在圖7C中，已經將耐韻遮罩720移除並加上一個耐蝕遮罩729來遮 覆η型區域。接著利用如第US5,648,277號專利所述的兩個加速麵來 植入雜質軒以提供p㈣性，使雜人的軒在整贿度上的濃度有更 均勻的分布。這會形成用來構成一個驅動電路的一個p通道型吓了的一 個源極區域730、-個没極區域73卜-個LDD區域732、和-個通道區 域733。雜質離子會透過一個退火程序使它活化。 接著形成-個第-中介層絕緣薄臈734並在其中開設多個接觸孔來 形成源極735、736、737和沒極738、739。絕緣層m可以使用從氧化 石夕、氮化石夕、氮氧化砍、和樹脂薄膜中挑選的材料製成。用於驅動電路的m 現在已經完成。接下來賴絲麟像練_ TFT。麵成祕和没極 之後，接著形成-個第二中介層絕緣薄膜⑽，然後在其上形成一個包括 Ti薄膜的黑色遮罩74卜如果我們在形成黑色遮罩741之前先在及極 挪上的位置部份移除第二中介層絕緣薄膜，則可以從黑色遮罩、第二中 介層絕緣薄膜、和沒極形成一個辅助電容。接著在黑色遮罩741上形成一 個第三絕緣層薄膜742並在其中形成—個接觸孔，以及在其上形成一個包 括有透明導電薄膜（諸如氧化銦錫）的像素電極⑽。 如圖7D中所示，這裡揭露的是—個包含τπ的主動矩陣式基板，它 包括有完整形成的像素和驅動電路，這些電路可能彼此緊鄰。熟悉這類技 術的人將會錢瞭觸7D㈣_魏可以取代域他魏，像是訊於 處理電路，賴㈣衫峨_塊。梅議區域_ 極關場效辦，陶偷軸“輕域相較於 88 200916986 多晶矽鍺區域具有較差的運算速度的特性，但多晶矽區域在應用於像素矩 陣TFT中時卻也具有較佳的低關閉（0FF)電流特性。 可以透過這種製造方法製成的構造並不限於TFT構造，而可以套用於 任何已知的構造，包括下閘極式TFT。 雷射光源 以（例如）GalnAs或GalnAsN材料為基礎的RGB固態雷射光源，由 於它們的精巧性質以及它們所具有的高水準光定向能力，可能非常適合用 來做為-個全像顯不H的絲。這—類的光源包括發光二極體以及由美國 加州的Novalux⑽）公司所製造的RGB垂直面射型雷射（VCSEL)。這 種雷射光財單束雷射或_雷射_式供應，賴每—種光騎可以透 過利用繞料光學TL件絲產生多道光束。光束可能會通過乡模光纖，因 j這可以t致性水準，如果—致性轉太高在騰精巧型全像顯示 =夺這可以避免產生令人不悅的人工處理痕跡，像是雷射光斑。雷射 光源的陣列可以是一維或二維的陣列。 基板 我們必須強調所謂的“基板，，也就是用來製造顯示器的一個材料板 塊。《常會是—個絕緣基板，像是玻璃板基板、或是—個藍寶石基板、 麵疋個半導體基板（如&amp;《㈣），但其他的基板像是聚合物板材 或金屬板材也是可行的。基板，如玻璃基板轉導體基板（像是&amp;或 89 200916986An acid resin, a sulphamine, a polyamidamine, or a sulphate is used as the eucalyptus membrane. Then, a black turn (10) is formed on the film. Weaving a third insulating interposer film 1162 having a thickness in the range of U 83 200916986 μm to 0.3 μm, such as oxidized oxidized stone, nitrided cerium, nitrous oxide oxide, or an organic resin film, or these materials Laminated film. Each of the contact holes ' is formed on the film and the film 1162 and a pixel electrode 1163 having a thickness of 12 Å is formed. An auxiliary capacitor 1164 is formed in a region where the black mask and the pixel electrode overlap, as shown in Fig. 13C. When the entire substrate is heated in a hydrogen-containing environment at a temperature of 35 (η: to 2 hours, this can provide compensation for the floating bonds, especially in the active layer of each film. After these steps, The CMOS circuit shown on the left side and the pixel matrix circuit shown on the right side of FIG. 13C are formed on the same substrate (for example, at adjacent positions). The PMOS circuit can be made by the manufacturing method and can be made thinner than the τπ structure. The sleeve is used in any known configuration, including a lower gate TFT. Third Manufacturing Procedure Overview In the basic structure of a thin film semiconductor display device of one display design of the present invention, a display member is provided in which the circuit component is disposed Displaying the components between the pixels of the display component or the other pixels on the same substrate, and performing various calculations related to the display device for displaying the data. The display component and the circuit component for performing the calculation on the wire, Peripheral, but the Jianerer component is used to operate the TFT circuit components of the spatial light modulator, and the other circuit components of the ', may be as follows Method cloth ^^ Method class rest, then 7 fine towels, 2 sets: 84 200916986 The method of making '· ·, 759, 677 patents are incorporated by reference in this document. Other methods are familiar with this type of technology. People do not know what to say. The semiconductor surface made by this manufacturing process is polycrystalline, and the electrical complexity of (4) may be similar (or even more than) to the single crystal in some respects (or in many respects). Electrical characteristics. This manufacturing process allows circuit components to be formed on the same substrate. Polysilicon can be used as a layer to generate a number of pixels that can be used to control the display. Other TFTs produced can have images. It is the function of _drive circuit, source-dynamic circuit, and signal processing circuit. In these TFTs, it is Shi Xi wrong, in order to improve the ability of high-speed operation. Ge will be added to the need for high-speed computing power. Part of the circuit component, and polycrystalline Si is used in the circuit part that needs to have a low (10) current. In this program, it will be manufactured—she has a pixel-to-circuit, a-drive circuit (which is a GMQS circuit), and these are all on the same substrate. — formed on an insulating surface. This processing procedure is as shown in Fig. 6. As shown in Fig. 6A, a glass substrate 6 () 1 is prepared, and then a layer of oxidized stone is formed thereon. An amorphous crystal film coffee having a thickness of 3Q nm was formed by a plasma CVD method. A mask #6〇4 was formed on the amorphous Si film 603 by a forming method. This refractory mask is formed to cover areas that will be used to form the scratch group for the pixel matrix circuit. The area that will be used to form the high speed circuit is not obscured. As shown in Fig. 6β, ^ will be implanted with smear, plasma doping, or "laser doping," the technique of adding seven enthalpy in order to change the composition of the amorphous Si chi to produce a uniform composition of sia thin 85 200916986, ... 1). If ion implantation technology is used, the film 605 to be added with Ge will be at risk of implant damage. Because π υ film coffee is in a non-crystalline In the state, the activation energy required for lattice diffusion in Ge is lower than that in Si, and Ge and Si will become mutual solid solutions at an alloy phase temperature below sub-hybrid temperature. Therefore, the presence of &amp; can be used to accelerate the crystallization of the Sh-xGex film (compared to the crystallization of only pure Si films). In IS ViEFi Γ J- , Ge is on the crystal of Si (like laser-induced crystallization) It can be considered as a catalyst semiconductor. In Figure 6C, the resist layer 6〇3 has been removed and a layer of Ni-containing layer 606 is applied over the entire surface as in U.S. Patent No. 5,643,826. (JS5, 643, 826 patents are incorporated by reference in this document Μ is used as a kind of catalyst material to force the mouth velocity & smear the crystal of Sll_xGeX film, but elements other than Ni, such as Co, Fe, Cu, Pd, Pt, Au, or In can also be used for this. The purpose is as shown in Fig. 6)), &amp; and % such as the crystallization of the film can be given to the Xu Lenghua Society to achieve, at the temperature of · C, about M, when: will form a polycrystalline Sil.xGex region 6〇7 and a polycrystalline stone area 608. This heat treatment can also be performed purely by other methods such as laser annealing or lamp annealing. As shown in Fig. 6E, the polycrystalline Sii xGex region 6〇7 will The active layer of the coffee towel is formed. The multi-day Si region 608 is formed in the active layer 61. The active layer is used as a TFT which will constitute a driving circuit and a signal processing circuit later; The active layer 610 is used as a working layer of a TFT which will constitute a pixel matrix circuit later. Then, the processing program described in US Pat. No. 5,648,277 is used to form a 86 200916986 source. Polar region, - polar region, and - her doping secret (ldd) region US5, _ patent The merging is used as a reference in this document. In the following, we will summarize this processing procedure. First, we use a film containing 2 wt% # Sc force ai to form an island shape that will be used later. Down, the anodization is performed on the island shape to form a film on the island ship. Then the solution is changed to perform the anodization to form an anodic oxide film surrounding the island. Shape: After forming a porous anodic oxide film and a compact anodic oxide film in a manner, a dry dielectric film is used for the side gate dielectric film. After the gate dielectric film is completed, the multi-recorded anode oxide is removed, thereby obtaining the energy shown in _a. As shown in Figure 7A, 712, and ? 13 is composed of a thin film of oxide oxide thin film 714, 715, 716 B m " her body closed, and 717 718 彳 719 is used to protect the gate of the compact anodic oxide film. As shown in 7B, the portion of the P-channel type m is covered with a cover. The remaining portion is subjected to n-type ion implantation to provide an n-type conductivity. Patent No. 5,648,277 In the above, there will be two non-accelerating voltages to make the concentration of the chisel rotor more uniform throughout the deep production. In Figure 7B, this process will form a drive circuit. A channel-type τπ--------------------------------------------------- A non-polar region 726, a source region, a (10) region 727, and a channel region 728. 87 200916986 In Figure 7C, the rhyme mask 720 has been removed and an anti-corrosion mask 729 is added to cover The n-type region is covered. The two acceleration faces as described in U.S. Patent No. 5,648,277 Implanted with impurities to provide p (four), so that the concentration of the miscellaneous people in the bribe is more evenly distributed. This will form a source region 730 that is used to form a drive circuit, a p-channel type scared - a non-polar region 73--LDD region 732, and a channel region 733. The impurity ions are activated by an annealing process. Then, a first-interposer insulating thin layer 734 is formed and a plurality of contact holes are formed therein. The source electrodes 735, 736, 737 and the gate electrodes 738, 739 are formed. The insulating layer m can be made of a material selected from the group consisting of oxidized stone, cerium nitride, nitrogen oxide chopping, and a resin film. Now that it has been completed. Next, Lai Silin is like _TFT. After the surface becomes secret and immersed, then a second interposer insulating film (10) is formed, and then a black mask 74 including a Ti film is formed thereon. Before the black mask 741 is formed, the second interposer insulating film is partially removed at the position of the pole mask, and an auxiliary capacitor can be formed from the black mask, the second interposer insulating film, and the immersion. In black mask 74 A third insulating layer film 742 is formed on the first insulating layer 742 and a contact hole is formed therein, and a pixel electrode (10) including a transparent conductive film such as indium tin oxide is formed thereon. As shown in FIG. 7D, it is disclosed herein. It is an active matrix substrate containing τπ, which includes fully formed pixels and driving circuits, which may be in close proximity to each other. Those who are familiar with this kind of technology will have to touch 7D (4) _ Wei can replace the domain Wei, like It is the processing circuit, Lai (four) shirt 峨 _ block. Mei 区域 area _ pole off field effect, Tao stealing axis "light domain compared to 88 200916986 polycrystalline 矽锗 region has poor operating speed characteristics, but polycrystalline 矽 region in the application It also has a preferred low off (OFF) current characteristic in the pixel matrix TFT. The construction which can be made by this manufacturing method is not limited to the TFT configuration, but can be applied to any known configuration including the lower gate TFT. Laser source RGB solid-state laser sources based on, for example, GalnAs or GalnAsN materials, due to their ingenious nature and their high level of light directivity, may be well suited for use as a holographic image. Silk. Light sources of this type include light-emitting diodes and RGB vertical plane-type lasers (VCSELs) manufactured by Novalux (10), California. This type of laser illuminating single beam laser or _laser _ type supply, by each type of light riding can be used to generate multiple beams by using the optical TL wire. The beam may pass through the rural mode fiber, because j can be t-level, if the temperament is too high in the dexterous full-image display = this can avoid unpleasant artificial processing traces, such as laser spots. The array of laser sources can be a one- or two-dimensional array. The substrate we must emphasize the so-called "substrate, that is, a material plate used to make the display. "It is often an insulating substrate, such as a glass substrate, or a sapphire substrate, a semiconductor substrate (such as &amp; "(4)), but other substrates such as polymer sheets or sheet metal are also possible. Substrates, such as glass substrate to conductor substrates (like &amp; or 89 200916986

GaAs)，經常會用在裝置的製造中，因為它們可以簡化處理步_及在執行 不同的製程步驟的各個設備（像是材料沈積、退火、以及材料触設傷)丁 之間的輸送。所_ “基板，，並不是指稱—個個別個電路板，如同 shim〇baba等人在0pticsExp職期刊（13，侧，襲）中所揭露的： -個個別電路減不允許在其上進行可財—_綱基板（像是玻璃板 基板）上執行的一系列製造程序。 電晶體總數估算 本節將_在-個全像計算將在佈置麵轉的各轉叙間的電路 組件上執行嶋示H巾所f的電晶體的數目的估算。 全像影像計算包括有下列的各個 上可以用於特定步驟的邏輯資源 在採用一個FPGA的顯示設計方面， 步驟，其中所示的百分比是表示在FPGA 的百分比。 •透鏡函數：加上隨機的相位並根據z值來產生次全像影像（㈣）GaAs) is often used in the manufacture of devices because they simplify the processing steps - and the transfer between individual devices (such as material deposition, annealing, and material contact damage) that perform different process steps. _ "Substrate, does not refer to a single circuit board, as shim〇baba et al. in the 0pticsExp job journal (13, side, attack) revealed: - individual circuit reduction is not allowed on it A series of manufacturing procedures performed on a substrate (such as a glass plate substrate). Estimation of the total number of transistors This section will perform a _in-total hologram calculation on the circuit components between the retellings of the layout surface. Estimation of the number of transistors in the H. The holographic image calculation includes the following logic resources that can be used for specific steps in the display design using an FPGA, the steps shown therein are expressed in the FPGA. Percentage. • Lens function: add random phase and generate sub-holographic image based on z-value ((4))

• C0RDIC計算：將取自她和度量的各個娜轉換騎麵虛值 執行光強度的調制（62.5%) W •疊加各個次全像影像來構成全像影像（15.5%) 罝以 •全像影像編碼：C_C運算法也會用來將各個值轉換為相位和度 及轉換回實值和虛值’顧於資料剪裁和正常化（Π.5%) 又 ，以上所示的 由於用於記齡元的電晶體數目並不取決於管線的頻率 200916986 百分比鮮錄行«矩料的運料可齡 的運算需求將會隨著全像影像的像素數增加而升高。 碼 像影和(LF)可能會有—些較小的LUT來定義取決於Z值的次全 7y 、大J、和透鏡函數的起始常數。所以绣铲 彳透鏡函數有—個相對較高的固 疋电日日體數目用於LUT以及一個敌'五认— 取决於母一個時鐘週期根據透鏡函數所 、Μτϋ動的C0RDIC單元數的可變電晶 體數目。—般而言，運算裝置（叢 集）的大小必須是最佳化的， — 1的大小越大，在資料傳輸速率中所 月匕產生的節約效果越小。但另一方 蕞集越大，將可以讓計算更容易實 現。圖23中的範例只顯示出—個簡 桃'Γ的叢集设计，因為-個叢集可能包 含有一百萬個電晶體或甚至更多。 現在，我們來估算-個全像計算將由佈置在顯示器的各個像素之間的 電路組件執行的顯示器所需要的電晶體數目。因為在_顯示設叶中 ^運_戀軸了5% _«，㈣梅算將集中在 用來執行C0RDIC計算的電晶體上。太 本文件中合併了參考文獻 C M_men ’ Architektumi _ 咖⑷級&amp;• C0RDIC calculation: converts the imaginary value of the riding surface from her and the metric to the modulation of the light intensity (62.5%) W • superimposes each hologram image to form a holographic image (15.5%) • • 全 影像Coding: The C_C algorithm is also used to convert each value into phase and degree and convert back to real and imaginary values. 'Data clipping and normalization (Π.5%) Again, as shown above for ageing The number of transistors in the element does not depend on the frequency of the pipeline. 200916986 Percentage of fresh recordings. The computational requirements for the age of the material will increase as the number of pixels in the hologram increases. The code umogram and (LF) may have some smaller LUTs to define the starting constants of the sub-full 7y, large J, and lens functions depending on the Z value. Therefore, the shovel lens function has a relatively high number of solid-state solar celestial bodies for the LUT and an enemy 'five---depending on the number of C0RDIC units that are pulsating according to the lens function and Μτ according to the mother's clock cycle. The number of transistors. In general, the size of the computing device (cluster) must be optimized, and the larger the size of 1, the smaller the savings in the data transmission rate. But the larger the other set, the easier it will be to make the calculations easier. The example in Figure 23 shows only a cluster design of a simple peach, because the clusters may contain one million transistors or even more. We now estimate the number of transistors required for a display to be performed by a circuit component disposed between the various pixels of the display. Because in the _ display set leaves _ _ love axis 5% _ «, (four) Mei will be concentrated on the transistor used to perform C0RDIC calculation. The reference is incorporated in this document. C M_men ’ Architektumi _ Coffee (4) Level &amp;

Realisierungen fflit Anwendungen in der BUdverarbeitung ^ D.rk T赚rraann’_〕做參考’摘錄其中的第⑽頁到第仙頁的内容， 來幫助估算c_c所需的電晶體數目。針對FPGA解決方案，已經開發 出使用有不同換算的-個改良的C0腿c單元，而據此所估算的用於一個 管線的C0RDIC單元的電晶體數目大約是52,〇〇〇電晶體。 圖21和22中所示的試算表顯示出針對所規劃的全像影像計算（有源 91 200916986 自-個2,咖x丨，個像素的真實空間影像的16,晒χ i2，_個全 像影像像素）的估算。次全像影像中的每—個像素都需要進行一個c〇腿c 運算，也就是每秒總計要進行灘1(Γ9個運算。以&amp; 的管線頻率 來算，需要有_舰_ _)IG單元。叢集的設計會雜電晶體的 數目和設計效能，因為叢集越大，表示用於分布全像影像資料的資源消耗 越多。但是如果叢集太小，在叢集中所進行的運算將會沒有效率，因為某 些單元將會大部分時間“無所事事，，而只是徒然增加電晶體數目而已。 如果一個叢集包括有1個透鏡函數單元和i個C0RDIC單元，將會 需要有9_個叢集和6億6千萬個電晶體來進行次全像影像運算。如 果一個叢集包括有1個透鏡函數單元和8個C0RDI(：單元，則顯示器將 會需要包括有1200個叢集和5億3千萬個電晶體用來進行次全像影像 運算。所以叢集的大小可能會有極大範圍的變化。針對我們的設計範例， 我們選擇一個叢集包含有4個CORDIC單元和1個透鏡函數單元。估算 的結果’這會需要有2500個叢集和5億5千萬個電晶體來進行次全像 影像運算。 要找出最理想的叢集大小，必須進行非常精細的設計。試算表中的數 子（圖21和22)只是一個大略的估算，但它們卻也顯示出各個參數的主 要依存形態。 CORDIC (位數對照方法，Voider運算法）（CORDIC也就是“座標轉 動數位電腦”的縮寫）是一種減單而效率極佳的運算法，可用來計算雙曲 線和三角函數。由於這裡是將CORDIC用來將來自相位和度量值的複雜數 92 200916986 值轉換為貫值和虛值fe e、 反向轉換），因此也可以採用其他的運算法。如 果沒有硬體倍增器（例如，簡單的微控制器和職）可用，通常就會採 用C0RDIC，因為它只需要小量的檢視表、位數移位、和加法計算。除此之 外’右在軟體或專用的硬體上執行，c_c運算法非常適合用於管線作 2現代的_IC運算法首先在年由Μ e.刪π提出儘 e匕矛Henry Briggs早在1624年所發表的技術非常類似。最初，⑽Ic 是以二進位制進行的，到了酬年代，十進位制⑽π已經廣泛應用 在口袋型計算機上’它們大部分都不是以二進_運算，而是以二進位編 碼十進位制（BCD)運算。⑽DIC特別適用於手持計算器，在這個應用方 面’它的成本（以及晶壯的閘極數目）雜它的速度更為重要。在沒有 硬體倍增ϋ (例如，微控中）可用時、或者在需要減知來執行所需 的閘極數目（例如，FPGA中）時’_IC的速度通常會比其他方法更快。 C0RDIC是“移位與疊加”類運算法的一部分，如同從Henry 的研究中導出的對數與指數運算法。另一種可以用來計算許多種基本函數 的移位與豐加運算法是BKM運算法，這是一種用於複數平面的廣義的對數 與指數運算法。舉例來說，BKM可以用來透過計算〇 + ix的指數（也就 是cosx+isinx)來計算一個實際角度X (弧度）的正弦和餘弦。在1994 年最先由 J.C. Bajard、S. Kla、和 J.M. Muller 在 IEEE 電腦會刊（43 (8): 955-963，1994年8月）中提出的BKM運算法比C0RDIC猶微更 複雜些，但它具有不須要使用一個比例因數的優點。BKM運算法在本發明 的顯示設計中可以用來代替C0RDIC運算法。 93 200916986 運算方法 今天，中央處理單元（CPU) 數位同步賴錢行縣。職全鮮彡料算柯 知用 4。由於每個全像影像像素所需的 知用唆個方法來運 可以採用其他的方法。以下、％晶體數目較低，根據運算的步驟，也 W—些其他的運算奴紅要屬性： 數位同步邏輯（時鐘邏輯） •適用於高電晶體數目 •運算時間短 *時序計算簡單 *有良好的設計工具支援 數位非同步邏輯（非時鐘邏輯） *有良好的功率效率 •適用於高電晶體數目 •運算時間短 •沒有良好的設計工具支援 •時序計算困難 PWM (脈衝寬度調變） 94 200916986 適用於低電晶體數目 •運算時間長 類比 •主要是在 S彳删年代發展出來 •除用於簡單的高頻用途外，類比«在今天已經不常用 .適用於極低的電晶體數目 •運鼻時間短 •精密度有限 •高度依賴生產參數飄移 混用多種技術 各個運异步驟的需求各不相同。由於諸如多㈣電晶體的運算容量有 限’運算方法娜據實際需麵藝。最麵方祕倾難㈣式設計 而疋。以下是一些範例。 為減少電晶體數目，有較低需求的運算步驟，像是透鏡函數和編碼， 可以採用酬。類比轉移暫存器可以用於資料分布，因為真實空間資料和 全像影像賴只會使關大約8位搞精密度。有—種特殊設計的非同牛 c_c單认可以絲減少功率消耗。每—個步驟種以上的方法還 可以進-步減少電晶體數目，不過這可能會增加設計的成本。 ^ 95 200916986 顯示器種類 ::使：顯示器最好是一種使用在顯示器一 凡如魏、先學開關）的主動矩陣式架構的顯示器。電晶體材料應有 -個適當的結構寬度以及開/關率來佈置額相電晶體賴運算使用。 可以使用單晶朴多晶_各種變化形態諸如低溫多⑽（聰）、⑽、 單體晶_、或多晶_。非結晶秒的開/關頻率—般來說對於高性能的 全像影像計算將會太低。上，有機半導體或奈米碳管也可簡來做為 開關元件的材料。傳統大型顯示器需要有極大的_絲容納橫列和縱列 線路，若採用本發明的方法將可以節省這樣的空間需求。 由於越大的顯示器上所能節省的空間越多，因此我們偏好採用下列型 式的顯示器： • LTPS式液晶顯示器（LCD) • LTPS式有機發光二極體（〇LED)〔包括發光聚合物（LEp)〕 單晶石夕只用於小型顯示器，相較於新的方法，它的優點較少。採用單 晶石夕範例包括：Realisierungen fflit Anwendungen in der BUdverarbeitung ^ D.rk T earns raraann’_] as a reference to extract the contents of page (10) to the first page to help estimate the number of transistors required for c_c. For the FPGA solution, a modified C0 leg c unit with different scaling has been developed, and the number of transistors estimated for the CORDDIC unit of one pipeline is approximately 52, 〇〇〇 transistor. The trial charts shown in Figures 21 and 22 show the calculations for the planned hologram image (active 91 200916986 from -2, coffee x 丨, pixel real image of the 16th, sunning i2, _ full Estimation like image pixels). Each pixel in the sub-holographic image needs to perform a c-leg c calculation, that is, a total of 1 per second (Γ9 operations. Calculated by the pipeline frequency of &amp; _ship__) IG unit. The design of the cluster will be the number of transistors and the design efficiency, because the larger the cluster, the more resources are consumed for distributing holographic image data. But if the cluster is too small, the operations performed in the cluster will be inefficient, because some units will spend most of their time "doing nothing, but just increasing the number of transistors in vain. If a cluster includes 1 lens function unit And i C0RDIC units, will need 9_ cluster and 660 million transistors for sub-holographic image operations. If a cluster includes 1 lens function unit and 8 C0RDI (: unit, then The display will need to include 1200 clusters and 530 million transistors for sub-holographic imaging operations. So the size of the cluster may vary greatly. For our design paradigm, we choose a cluster to contain There are 4 CORDIC units and 1 lens function unit. The estimated result 'This will require 2500 clusters and 550 million transistors for sub-holographic imaging operations. To find the optimal cluster size, you must do Very fine design. The numbers in the spreadsheet (Figures 21 and 22) are only a rough estimate, but they also show the main dependencies of the various parameters. CORDIC (Voider Control Method, Voider Algorithm) (CORDIC is also the abbreviation for "coordinate rotary digital computer") is an efficient and efficient algorithm for calculating hyperbolic and trigonometric functions. Since this is CORDIC It is used to convert the complex number 92 200916986 values from phase and metric values into the traverse and imaginary values fe e, reverse conversion), so other algorithms can be used. If there is no hardware multiplier (for example, simple micro Controller and job) are available, usually C0RDIC, because it only requires a small amount of view, bit shift, and addition calculation. In addition to 'right on software or dedicated hardware, c_c operation The method is very suitable for pipelines. 2 Modern _IC algorithm Firstly, the technique published by Henry Briggs as early as 1624 is very similar in the year. In the age of pay, the decimal system (10) π has been widely used in pocket computers. 'Most of them are not binary operations, but binary binary decoding (BCD) operations. (10) DI C is especially suitable for handheld calculators, where the cost (and the number of gates in the crystal) is more important in this application. When no hardware multiplication (for example, in micro control) is available, or The speed of '_IC is usually faster than other methods when it is necessary to reduce the number of gates required (for example, in an FPGA). C0RDIC is part of the "shift and superposition" class algorithm, as studied from Henry The logarithm and exponential algorithm derived in . The other shift and augmentation algorithm that can be used to calculate many kinds of basic functions is the BKM algorithm, which is a generalized logarithm and exponential algorithm for complex planes. Say, BKM can be used to calculate the sine and cosine of an actual angle X (radian) by calculating the 〇 + ix index (ie cosx + isinx). The BKM algorithm first proposed by JC Bajard, S. Kla, and JM Muller in the IEEE Computer Journal in 1994 (43 (8): 955-963, August 1994) is more complicated than the C0RDIC. But it has the advantage of not having to use a scaling factor. The BKM algorithm can be used in place of the C0RDIC algorithm in the display design of the present invention. 93 200916986 Algorithm Today, the central processing unit (CPU) is digitally synchronized. The job is full of fresh oysters. Other methods can be used because of the knowledge required by each holographic image pixel. The following, the number of % crystals is low, according to the steps of the operation, also W - some other operations slave red attributes: digital synchronization logic (clock logic) • for high transistor number • short operation time * simple timing calculation * good Design tools support digital non-synchronous logic (non-clock logic) * Good power efficiency • Suitable for high transistor counts • Short computation time • No good design tool support • Timing calculations difficult PWM (Pulse Width Modulation) 94 200916986 Applicable to the number of low-transistors • Long-term analogy of operation • Mainly developed in the era of S • • In addition to being used for simple high-frequency applications, the analogy «is not commonly used today. Suitable for extremely low number of transistors. Short nose time • Limited precision • Highly dependent on production parameters drifting and mixing of various technologies The requirements for each of the different steps are different. Since the computational capacity of a multi-(tetra) transistor is limited, the calculation method is actually required. The most difficult part of the secret is the (four) design and the embarrassment. Here are some examples. In order to reduce the number of transistors, there are lower-cost computing steps, such as lens functions and encoding, which can be used. The analog transfer register can be used for data distribution, because real-world data and full-image images only make about 8 bits of precision. There is a special design of the non-conformity c_c single recognition can reduce the power consumption. The method of each step or more can further reduce the number of transistors, but this may increase the cost of the design. ^ 95 200916986 Types of Displays :: Make: The display is preferably a display of an active matrix architecture used in displays such as Wei and Xianxue. The transistor material should have an appropriate structure width and an on/off rate to place the front-phase transistor. Single crystal polycrystals can be used - various variations such as low temperature (10) (sat), (10), monomer crystals, or polycrystalline. The on/off frequency of amorphous seconds is generally too low for high performance holographic image calculations. Upper organic semiconductors or carbon nanotubes can also be used as materials for switching elements. Conventional large displays require a very large number of wires to accommodate the course and column lines, which would save space requirements if the method of the present invention is employed. Since the larger the space saved on the display, the more we prefer to use the following types of displays: • LTPS liquid crystal display (LCD) • LTPS organic light emitting diode (〇LED) [including luminescent polymer (LEp) )] Single crystal stone is only used for small displays, and it has fewer advantages than the new method. Examples of using the monocrystalline eve include:

• LC0S •數位光處理（DLP)技術 可以用於本發明的一個顯示設計的可能顯示技術名單如下： 液晶顯示器（LCD)—類別 LC0S矽液晶 96 200916986 NLC 向列型液晶 TN 扭轉向列型 VAN 垂直排列向列型 FLC 強誘電性液晶 FED (場發射顯示器） SED表面傳導電子發射顯示器 奈米礙管發㈣（以雜板或塗佈氧化轉（圈的賴基板為基 礎，但這些只能用來做為光源，因為它會發射非一致性的光） 電機系統 反射鏡陣列/數位光處理（DLP)技術 MEMS反射鏡（微電機系統）’也稱為M〇EMS (微光電電機系統） 全像影像計算的方法包括： -檢視表（LUT) -分析式運算 一第W0 2006/066919號專利公告中所述的方法，合並於本文件中做為參 考。 ' -光線追蹤方法 97 200916986 轉換類別： -2D轉換 -水平面中的1D轉換 _垂直面中的1D轉換 編碼類別： -Burckhardt 編碼 -唯相位編碼 -雙相位編碼 -BIAS編碼 -瞧（最小距離編碼）—每個全像像素使用3個以上的灿像素的編 硬體 一個外部全像影像計算單元，可能包括—對高階腸或包含有約5 億2千萬個電晶體及· MHz的管線頻率的—個根據應職定的積體電 路（ASIC)或-個純客製積體電路⑽。為傳送大約23〇對低電壓差分 訊號⑽s)的資料給顯示器，每一對的傳輪可以使用每秒i⑶加的^ 率。為接收資料，也會需要玻璃覆晶（C〇G)橫列及縱列線路驅動裝置。如 果運算將^整合在顯7^龜上’則高m頻料料諸域位視覺介 面（Wi)接收器必須佈置在額外的硬體上。必須只能以低5〇倍的資科傳 98 200916986 輸速率來傳送原始資料（請參閱圖D。可以使用與顯示器只有少數連接線 路的低{貝位顯不器電子系統。這種電子系統大概類似於今天的低解析度2D TFT顯示器中的電子系統。 附註： U上二節概略敘述了可能結合的製造方法而沒有偏離本發明的範圍的 . 特點。 〔 在本文件的附圖中，所顯示的相關尺寸並不必然符合比例。 對於戒悉賴技躺人來說且，本發明的各種修改及變更將是顯而易 見的1_仍不會超&amp;本發明的範圍，賴悉這類技術的人應可瞭解本發明 並未過度限定於本文件所附的圖解範例。 j文件中包含有多種概念（如“概念A — T”所述）。附錄ΠΙ包含 有可此有助於界定③些概念的内容。熟悉這類技術的人將會非常清楚，揭 ^個概念可能將有助於_其他的絲。本文件中的某些齡可能構成 v +可或缺的—部分’並且將在本文件的其他地方有清楚的闡述。 99 200916986 附錄. 技術入門 以下這一節將用來做爲多項用於某些套用本發明的系統中的關鍵技術的基本介紹。 在傳統全像技術中’觀察者可以看到一個物件的全像影像重建（這可能是一個不斷變動 的景象）；不過’觀察者與全像影像之間的距離可能無關。這種重建，就典型的視覺配 置來說，是在照亮全像影像的光源的影像平面上或它的附近，也就是在全像影像的傅立 葉平面上或它的附近。因此，這個重建影像有與被重建的真實世界物件相同的遠場光分 布。 —個早期的系統（如第WO 2〇04/044659及US 2006/0055994號專利中所述，整體合 並在本文件中做爲參考）定義了一種非常不一樣的配置，其中重建的物件完全不在全像 影像的傅立葉平面上或它的附近，相反的，貝IJ是一個虛擬觀察者視窗區域在這個全像影 像的傅立葉平面上；觀察者將眼睛放在這個位置上，並且將只能看到一個正確重建的影 像。全像影像會在一個LCD (或者其他類型的空間光調制器）上編碼，並透過一個光 學設計照亮，使全像影像的傅立葉轉換就在虛擬觀察者視窗上進行（因此也就是一個傅 立葉轉換値接映射到眼睛上）；在觀察者視窗與SLM之間的一個截頭錐體空間中形成 的重建物件之後則透過全像影像的菲涅爾轉換在傳播上做更佳的描述，因爲它並不是在 透鏡的焦點平面上。相反的，它是由一個近場光分布所界定的（利用球面波前構形，與 —個遠場分布的平面波前相反）。這個重建可以出現在虛擬觀察者視窗（如前面所述， 也就是全像影像的傅立葉平面）與SLM之間的任何地方或甚至可以出現在SLM後面 成爲一個虛像物件。 這種方法會帶來多種後果。首先，對全像視訊系統設計者的基本限制是SLM (或者其 他類型的光調制器）的像素間距的問題。目標是可以使用能以合理的成本從市面上購得 的適當像素間距的SLM來進行夠大的全像影像重建。但在過去，這卻由於下列因素而 無法實現。傅立葉平面中的相鄰繞射級之間的週期間隔以λ/)/;?來表示，其中λ是照 100 200916986 明光的波長，乃是從全像影像到傅立葉平面的距離，而；？是SLM的像素間距。但在 傳統的全像顯示器上，重建的物件會出現在傅立葉平面上或它的附件。因此，一個重建 的物件必須保持小於週期間隔；如果較大，則它的邊緣將會因爲一個相鄰的繞射級的 重建而模糊。這會使重建的物件變得非常小一通常寬幅只有幾cm，即使是使用所費 不貲的專用小間距顯示器也是一樣。但若採用本發明的方法，虛擬觀察者視窗（如以上 鎖數，這個視窗的位置就在全像影像的傅立葉平面上）的大小只需要像眼睛的瞳孔一樣 大即可。這樣的結果是，即使只有中等間距大小的SLM也能夠使用。同時因爲重建的 物件可以整個充滿虛擬觀察者視窗與全像影像之間的截頭錐體空間，因此它確實夠大， 當然也遠大於週期間隔。除此之外，如果使用一個OASLM，貝IJ不會有鋸齒狀的現象， 因此也週期性的問題，所以也不會再有虛擬觀察者視窗必須保持小於一個週期間隔的限 制。 若採用另一種變化方式，這還有另外一個好處。在計算一個全像影像時，我們先從我們 對重建的物件的認識說起一例如，您可能有一輛賽車的一個3D影像檔案。這個檔案 將會說明這個物件應該從哪些不同的觀看位置才能看見。在傳統的全像技術中，全像影 像必須在一個需要密集運算的過程中產生一個直接取自3D影像檔案中的賽車的一個 重建。但虛擬觀察者視窗的方法則可以採用一個不同的且運算效率更佳的技術。從重建 的物件的一個平面開始，我們可以以物件的菲涅爾轉換的方式來計算虛擬觀察者視窗。 我們接著可以針對所有的物件平面執行這個動作，並加總所有的結果來產生一個累計的 菲涅爾轉換；這會界定整個虛擬觀察者視窗的波場。我們接著再以這個虛擬觀察者視窗 的傅立葉轉換的方式來計算全像影像。由於虛擬觀察者視窗包含有物件的所有資訊，只 有單一平面的虛擬觀察者視窗（而不是多平面的物件）必須進行傅立葉轉換成爲全像影 像。如果從虛擬觀察者視窗到全像影像不是只採用一個轉換步驟而是採用一個反覆的轉 換步驟（像是反覆傅立葉轉換運算法），這會特別有利。如果需要迭代運算，則每一個 迭代步驟只包括有一個虛擬觀察者視窗的傅立葉轉換而不是每一個物件平面的傅立葉 轉換，這會顯著的減少運算的需求。 虛擬觀察者視窗方法的另一個有利的結果是重建一個特定物件點所需的所有資訊是包 含在全像影像中的一個相對小的部份中；這與傳統的全像技術不一樣（它重建一個特定 101 200916986 物件點所需的資訊是分散在整個全像影像的各處）。由於我們只需要將這些資訊編碼成 爲全像影像中的一個明顯更小的部份，這表示我們需要處理和編碼的資訊量遠少於一個 傳統的全像影像。這同樣也表示即使是即時的視訊全像技術，也可以使用傳統的運算裝 置（例如，成本和性能適合用於量產市場化裝置的一個傳統的DSP)。 不過，它也有一些可能不盡如人意的結果。首先，相對於全像影像的觀看距離將非常重 要一全像影像是以只有在眼睛處於全像影像的傅立葉平面上或附近的位置時才能看 見正確的重建的方式來進行編碼和投射的；而在一般的全像影像方面，觀看距離並不是 一個重要的因素。不過，已經各種不同的技術可以用來減少Z軸距離的敏感度或環繞 這個因素的設計困難度。 同時，由於全像景多像是以encoded and illuminated in such a way that正確的全像影像重 建只能從一個精密且極小的觀看位置（尤其是針對側向定位以及在Z軸距離而言）上 看見的方式進行編碼和投射的，因此可能需要進行眼睛追蹤。由於對Z軸距離極爲敏 感’因此需要採用各種技術來減少X、Y方位的敏感度或環繞這個因素的設計困難度。 例如’隨著像素間距的縮小（因爲它具有SLM在製造上的優勢），虛擬觀察者視窗的 大小將可以增大。除此之外，更有效率的編碼技術（像是Kinoform編碼）也有利於使 用一個更大的週期間隔的部份來做爲虛擬觀察者視窗，並因此可以增大虛擬觀察者視 窗。 以上的說明的一個前提是假設我們所麵的是傅立葉全像影像。麵觀察者視窗是在全 像影像的傅立葉平面上，也就是在光源的影像平面上。它的優點是，非繞射的光會聚焦 在所謂的DC點上。如果虛麵察者纖不是在光源的影像平社，這雛馳可以 用在菲涅爾全像影像上。不過，必須注意由於一個擾動的背景，非繞射的光是看不見的。 必須注意鹏-赔所鋼“轉換”麟觀包雛何麵於颇似棚賴明光的 讎的轉換的數學或運算上的技術。麵麵（Maxwellian)波讎等式對於僅只近似 於物理過程的轉換有更精確的定義;菲麵麵立麵酬是第二級近似，但它具有下 列讎:(a)因爲它們與微分相反的代數，所以可以以一個在運算上更有效率的方式來 處理’以及（ii)可以在光學系統中精確的實施。 102 200916986 進一步的細節可參閱美國專利申請案第US 2006-0138711、US 2006-0139710、及US 2006-0250671號（本文件合倂相關內容做爲參考）。 200916986 mm ii 本文件中所採用的用語說明 電腦產生全像影像 一個電腦產生的全像影像(CGH )是根據一個景象所計算出來的一個全像影像。CGH可 能包括代表用來重建景象所需的光波的振幅和相位的複合値數値。CGH可以透過計算 得出，例如，可透過“一致光線追蹤”、透過模擬景象與一個參考光波之間的干涉、或 者透過傅立葉或菲涅爾轉換。 編碼 編碼是一個程序，在這個程序中會提供全像影像的控制値給一個空間光調制器（例如， 構成它的像素格、或者一個連續的SLM如OASLM的相鄰區域）。一般而言，一個全 像影像包括有代表振幅和相位的複合値數値。 編碼區域 編碼區域通常是指全像影像中的一個限定空間的區域，在這個區域中會對一個單一的景 象點的全像影像資訊進行編碼。空間上的限制可能藉由一個突然的截斷、或者藉由透過 一個從虛擬觀察者視窗到全像影像的傅立葉轉換所達成的平滑截斷來實現。 傅立葉轉換 傅立葉轉換會用來計算光在空間光調制器的遠場中的傳播。波前則以平面波來描述。 傅立葉平面 傅立葉平面包含光在空間光調制器上的分布的傅立葉轉換。若沒有任何有焦點的透鏡， 傅立葉平面會在無限遠的位置。如果在光的行進路線中有一個有焦點的透鏡靠近空間光 調制器，則傅立葉平面相當於包含有光源的影像的平面。 菲涅爾轉換 菲涅爾轉換會用來計算光在空間光調制器的近場中的傳播。波前則以球面波來描述。光 波的相位因數包括有一個取決於橫向座標平方的二次項。 104 200916986 截頭錐體 虛擬觀察者視窗與SLM之間會構成一個虛擬的截頭錐體並一直延伸到SLM後方。景 象會在這個截頭錐體中重建。所重建的景象的大小會受這個截頭錐體限制而不會受 SLM的週期間隔限制。 光源系統 光源系統可能包括有一個一致的光源（如雷射）或一個部份一致的光源（如LED)。部 份一致的光源在時間上和空間上的一致性必須足以有利於進行良好的景象重建’發射表 . 面的也就是光譜線寬度和側向廣度必須夠/]、。 虛擬觀察者視窗（vow) ‘ 虛擬觀察者視窗是在觀察者平面上的一個虛擬視窗’透過這個可以看見經過重建的3D 物件。vow是全像影像的傅立葉轉換，並且位於一個週期間隔範圍內以避免看見物件 的多個重建影像。vow的大小必須至少是眼睛瞳孔的大小。如果至少有一個VOW設 置在觀察者眼睛的位置處並且具備有一個觀察者追蹤系統，則vow比觀察者側向移 動的範圍更小。這可方便使用一個只具有適當解析度（因此也具有較小的週期間隔）的 SLM。可以將VOW想像爲一個鑰匙孔，透過這個鏡匙孔我們可以看見重建的3D物 件；可以是每個眼睛使用一個VOW或者是兩個眼睛共用一個VOW。 週期間隔 如果CGH顯示在一個由個別的可編址像素格所組成的SLM上，會對它進行取樣。這 個取樣的動作會產生一個繞射圖樣的週期性重複。週期間隔以AD/p表示，其中λ是 波長，D是從全像影像到傅立葉平面的距離，而ρ是SLM像素格的間距。不過 OASLM並沒有進行取樣，因此繞射圖樣也沒有週期性的重複；這個重]複事實上是被抑 制的。 重建 將有全像影像編碼的空間光調制器照亮會重建原來的光的分布。這個光的分布會用來計 算全像影像。理論上，觀察者將無法分辨重建的光分布和原來的光分布。在大部分的全 像顯示器上’所重建的是景象的光的分布；而在我們的顯示器上，則是重建虛擬觀察者 105 200916986 視窗中的光的分布。 景象 要進行重建的景象是一個真實的或者由電腦產生的三維的光分布。在一個特殊的例子 中，它也可能是一個二維的光分布。一個景象可能包括有排列在一個空間中的各個不同 的固定或活動的物件。 空間光調制器（SLM) SLM會用來調制輸入的光的波前。一個理想的SLM將可以代表任意的複合値數値， 亦良P，可以分別控制一個輸入的光波的振幅及相位。不過，一個典型的SLM則只會控 * 制一個性質，不是振幅就是相位’而且還會有也可能會影響另一個性質不良副作用。 106 200916986 附錄 槪念 本文件中包含有多種槪念（如“槪念A〜T”所述）。以下的說明可能有助於界定這些 槪念。 A.可在和像素所在的同一基板上進行計算的全像影像顯示器 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行的全像顯示器。 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於空間光調制器的 各個像素之間的電路組件來執行。 .計算會在分散在顯示器中個各個區域執行’以便針對各個分散區域，進行各個相應 的分散區域的像素編碼。 •電路組件中包括有薄膜電晶體。 .至少有一部份電路組件的有效區位採用有多晶矽。 •至少有一部份電路組件的有效區位採用有連續晶粒矽。 •至少有一部份電路組件的有效區位採用有多晶砂鍺。 •至少有一部份電路組件的有效區位採用有單晶矽。 •至少有一部份電路組件的有效區位採用有單體晶粒矽。 107 200916986 •至少有一部份電路組件的有效區位採用有有機半導體。 •採用單晶矽基板。 .採用玻璃基板。 .只有真實空間影像資料會被傳輸給顯示器。 •視訊的影格速率至少約爲25 Hz。 •影像資料中包含光強度及緩衝區對應資料。 •所進行的全像計算爲即時或近即時運算。 •所進行的全像計算將採用檢視表方法執行。 •採用次全像影像來進行運算。 •用於加入次全像影像的資料會在一個次全像影像尺度的距離上進行交換。 •全像運算會均質的分散在整顯示器表面上。 •全像計算會分割爲拼貼在顯示器表面上的許多全等的微小部份（稱爲叢集）。 •用於加入次全像影像的資料會在一個叢集尺度的距離上進行交換。 •全像顯示器可以透過將許多全等的叢集拼貼在一起而構成。 108 200916986 .根據上述構想，本案之全像顯示器是一個高解析度顯示器。 •根據上述構想，本案之全像顯示器是一個極高解析度顯示器。 • 一個虛擬觀察者視窗的直徑大約是眼睛瞳孔的直徑或更大的直徑。 •這個虛擬觀察者視窗的直徑大約是1 cm或更大的直徑。 •會針對母一個眼睛（亦即針對每一個虛擬觀察者視窗）組織一組成對的緩衝區對應 及光強度對應資料。 •將可以顯示單色影像。 •將可以顯示彩色影像。 •將可以顯示RGB格式的彩色影像。 •在計算一個全像影像的一個像素的値時，只會考慮原始影像的一個次區段的値。 % •用於進行全像影像重建的光並非在整個顯示器上完全一致，而是在顯示器的各個次 區段範圍內完全一致。 .只需比傳輸全像影像資料所需更少的線路即足以用來傳輸原始影像資料。 •減低資料傳輸頻率具有可減少橫列和縱列驅動裝置中的功率消耗的好處。 •在先前的專利的解決方案中因縱列和橫列線路所需的大部分像素面積可以運用於其 他目的。 109 200916986 •透明電極的面積可以增加，並可因此而改善顯示器的透光度。 •顯示面板可以利用傳統顯示技術來控制。 •顯示器在製造上將採用矽晶液晶技術。 •顯示器在製造上將採用MEMS技術。 •顯示器在製造上將採用場發射顯示器（FED)技術。 •全像轉換爲一維轉換。 •全像轉換爲二維轉換。 •有一個額外的邏輯可用於區域轉送現有計算所得的資料，而這個額外的邏輯也同時 用於轉送原始影像給各個叢集，所以至少可以剔除某些共通的橫列和縱列線路。 .備用電路組件（像是TFT)可能會被製造在像素矩陣的空間中，使這些電路組件可 以在發現某些用於裝置啓動的電路組件有故障時，用來取代某些在裝置啓動時會用到的 電路組件。 •使用全像顯示器的方法。 B.可在同一基板上進行計算、可進行有效率的空間光調制器編碼計算的全像影像顯示 器 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行的全像顯示器，且這些計算的本身並不牽涉到 傅立葉轉換或菲涅爾轉換的計算。 110 200916986 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於空間光調制器的 各個像素之間的電路組件來執行。 •計算會在分散在顯示器中個各個區域執行，以便針對各個分散區域，進行各個相應 的分散區域的像素編碼。 •電路組件中包括有薄膜電晶體。 •至少有一部份電路組件的有效區位採用有多晶矽。 .至少有一部份電路組件的有效區位採用有連續晶粒矽。 •至少有一部份電路組件的有效區位採用有多晶矽鍺。 .至少有一部份電路組件的有效區位採用有單晶矽。 •至少有一部份電路組件的有效區位採用有單體晶粒矽。 •至少有一部份電路組件的有效區位採用有有機半導體。 •採用單晶矽基板。 .採用玻璃基板。 .只有真實空間影像資料會被傳輸給顯示器。 •視訊的影格速率至少約爲25 Hz。 111 200916986 •影像資料中包含光強度及緩衝區對應資料。 •所進行的全像計算爲即時或近即時運算。 •所進行的全像計算將採用檢視表方法執行。 •採用次全像影像來進行運算。 •全像運算會均質的分散在整顯示器表面上。 •全像計算會分割爲拼貼在顯示器表面上的許多全等的微小部份（稱爲叢集）。 •根據上述構想，本案之全像顯示器是一個高解析度顯示器。 •一個虛擬觀察者視窗的直徑大約是眼睛瞳孔的直徑或更大的直徑。 •將可以顯示單色影像。 •將可以顯示彩色影像。 •在計算一個全像影像的一個像素的値時，只會考慮原始影像的一個次區段的値。 •用於進行全像影像重建的光並非在整個顯示器上完全一致，而是在顯示器的各個次 區段範圍內完全一致。 •只需比傳輸全像影像資料所需更少的線路即足以用來傳輸原始影像資料。 •減低資料傳輸頻率具有可減少橫列和縱列驅動裝置中的功率消耗的好處。 112 200916986 •在先前的專利的解決方案中因縱列和橫列線路所需的大部分像素面積可以運用於其 他目的。 •透明電極的面積可以增加，並可因此而改善顯示器的透光度。 •顯示面板可以利用傳統顯示技術來控制。 •顯示器在製造上將採用矽晶液晶技術。 •顯示器在製造上將採用MEMS技術。 •顯示器在製造上將採用場發射顯示器（FED)技術。 •全像轉換爲一維轉換。 .全像轉換爲二維轉換。 •有一個額外的邏輯可用於區域轉送現有計算所得的資料，而這個額外的邏輯也同時 用於轉送原始影像給各個叢集，所以至少可以剔除某些共通的橫列和縱列線路。 •備用電路組件（像是TFT)可能會被製造在像素矩陣的空間中，使這些電路組件可 以在發現某些用於裝置啓動的電路組件有故障時，用來取代某些在裝置啓動時會用到的 電路組件。 •將由物件所發出的波前將會在一個或多個虛擬觀察者視窗（VOW)中重建，且其中 一個三維景象（3D S)的每個單一物件點（OP)的重建只需要一個次全像影像（SH) 做爲要在SLM上進行編碼的整個全像影像（PESLM)的子集。 •在一個景象（3D S)離散化成爲多個物件點（OP)之後，針對3D景象的每個可見 113 200916986 的物件點，透鏡次全像影像（SHL)的複雜値會在SLM上進行編碼，其中透鏡次全像 影像的複雜値會利用下列方程式來決定：ZL = eXp { -i*〔（W ) * ( X2 + y2川，其 中λ是參考波長’ f是焦距’而X和y貝隄在次全像影像的平面中互相垂直的座標。 •稜鏡的次全像影像（SHP)會在全像影像平面（HE)中決定以便移動虛擬觀察者視 窗以遠離光軸。 •透鏡和稜鏡的是次全像影像是迴旋的，這也可以用SH = SHL * SHP來表示。 •每一個次全像影像（SH)會以一個統一分布的相位偏移來進行調制，這裡的相位偏 移每一個次全像影像都不盡相同。 •會將多個次全像影像疊加在一起來構成整個全像影像。 •用於進行重建的電腦產生全像影像的呈現可即時或近即時更新。 •在全像計算中會使用檢視表。 •各個物件點可以在用來進行重建的截頭錐體空間中的任ί可位置產生。 •使用全像顯示器的方法。 C-可在同一基板上進行解壓縮計算的全像影像顯示器 全像影像編碼資料是在像素矩陣所佔用的空間以外的地方計算，然後在將全像影像編碼 資料利用已知的資料壓縮技術進行壓縮，接著再傳輸到顯示器基板上的電路組件，這些 電路組件接著再對所接收到的資料執行解壓縮功能的全像顯示器。 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器 114 200916986 的像素所在的同一基板上的電路組件來執行。 •電路組件中包括有薄膜電晶體。 .至少有一部份電路組件的有效區位採用有多晶矽。 •至少有一部份電路組件的有效區位採用有連續晶粒矽。 •至少有一部份電路組件的有效區位採用有多晶矽鍺。 •至少有一部份電路組件的有效區位採用有單晶矽。 •至少有一部份電路組件的有效區位採用有單體晶粒矽。 •至少有一部份電路組件的有效區位採用有有機半導體。 .採用單晶矽基板。 .採用玻璃基板。 •視訊的影格速率至少約爲25 Hz。 •影像資料中包含光強度及緩衝區對應資料。 •所進行的全像計算爲即時或近即時運算。 •所進行的全像計算將採用檢視表方法執行。 .採用次全像影像來進行運算。 115 200916986 •根據上述構想，本案之全像顯示器是一個高解析度顯示器。 •一個虛擬觀察者視窗的直徑大約是眼睛瞳孔的直徑或更大的直徑。 •將可以顯示單色影像。 •將可以顯示彩色影像。 •在計算一個全像影像的一個像素的値時，只會考慮原始影像的一個次區段的値。 •用於進行全像影像重建的光並非在整個顯示器上完全一致，而是在顯示器的各個次 區段範圍內完全一致。 •減低資料傳輸頻率具有可減少橫列和縱列驅動裝置中的功率消耗的好處。 •在先前的專利的解決方案中因縱列和橫列線路所需的大部分像素面積可以運用於其 他目的。 •透明電極的面積可以增加，並可因此而改善顯示器的透光度。 •顯示面板可以利用傳統顯示技術來控制。 •顯示器在製造上將採用矽晶液晶技術。 •顯示器在製造上將採用MEMS技術。 •顯示器在製造上將採用場發射顯示器（FED)技術。 116 200916986 •全像轉換爲一維轉換。 •全像轉換爲二維轉換。 •備用電路組件（像是TFT)可能會被製造在像素矩陣的空間中，使這些電路組件可 以在發現某些用於裝置啓動的電路組件有故障時，用來取代某些在裝置啓動時會用到的 電路組件。 • *將由物件所發出的波前將會在一個或多個虛擬觀察者視窗（v〇w)中重建，且其中 一個三維景象（3D S)的每個單一物件點（OP)的重建只需要一個次全像影像（SH) 做爲要在SLM上進行編碼的整個全像影像（IESLM)的子集。 •在一個景象（3D S)離散化成爲多個物件點（OP)之後，針對3D景象的每個可見 的物件點，透鏡次全像影像（SH。的複雜値會在SLM上進行編碼’其中透鏡次全像 影像的複雜値會利用下列方程式來決定：ZL = exp { -i*〔（π/λί ) * ( x2 + y2 )〕}，其 中λ是參考波長，f是焦距，而x和y則是在次全像影像的平面中互相垂直的座標。 •稜鏡的次全像影像（SHP )會在全像影像平面（HE )中決定以便移動虛擬觀察者視 窗以遠離光軸。 K. •透鏡和稜鏡的是次全像影像是迴旋的，這也可以用SH = SHl * SHP來表示。 .用來執行全像計算的空間可能是或可能不是在與顯示器基板所在的同一基板上。 .鋪執行解壓縮計算的電路組件位讎示器的各讎素之間。 •負麵行麵縮計算的電路組件位讎示器的像素矩陣以外、但在同—麵上的位 置。 117 200916986 •叢集會執行解壓縮計算。 •進行解壓縮計算的叢集會經由顯示器的橫列和縱列線路來接收資料。 •每一個進行解壓縮計算的叢集會經由一個並列資料匯流排來接收資料。 •每一個進行解壓縮計算的叢集會經由一個序列資料連接線路來接收資料。 - •使用全像顯示器的方法。 D.可在同一基板進行解壓縮計算的高解析度顯示器 一個高解析度顯示器會用來顯示高解析度影像資料，這些資料會先利用已知的資料壓縮 技術進行壓縮，然後再傳輸到顯示器的基板上的電路組件，這些電路組件接著再對所接 收到的資料執行解壓縮功能，之後再將資料顯示在顯示器的各個像素上。 •進行解壓縮的電路組件位於顯示器的各個像素之間。 •進行解壓縮的電路組件位於顯示器的像素矩陣之外、但在顯示器所在的同一基板上。 \ . ' •壓縮過的資料會傳輸到顯示器上構成整個顯示器的各個叢集上，叢集接著再對所接 收到的資料執行解壓縮功能，之後再將資料顯示在區域叢集的各個像素上。 •可以顯示一般顯示資料。 •可以顯示全像顯示資料。 •執行壓縮計算的空間可能在或可能不在與顯示器的基板所在的同一基板上。 118 200916986 •進行解壓縮計算的叢集會經由顯示器的橫列和縱列線路來接收資料。 •每一個進行解壓縮計算的叢集會經由一個並列資料匯流排來接收資料。 •每一個進行解壓縮計算的叢集會經由一個序列資料連接線路來接收資料。 •是一個極高解析度的顯示器。 - •每一個叢集會在40 ms或以下的時間內執行解壓縮。 .全像影像計算會在解壓縮之後執行。 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器 的像素所在的同一基板上的電路組件來執行。 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器 的像素所在的同一基板上的電路組件來執行，且這些計算的本身並不牽涉到傅立葉轉換 或菲涅爾轉換的計算。 i'•至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於空間光調制器的 各個像素之間的電路組件來執行。 •計算會在分散在顯示器中個各個區域執行，以便針對各個分散區域，進行各個相應 的分散區域的像素編碼。 •電路組件中包括有薄膜電晶體。 •至少有一部份電路組件的有效區位採用有多晶矽。 119 200916986 •至少有一部份電路組件的有效區位採用有連續晶粒矽。 •至少有一部份電路組件的有效區位採用有多晶矽鍺。 •至少有一部份電路組件的有效區位採用有單晶矽。 •至少有一部份電路組件的有效區位採用有單體晶粒矽。 - •至少有一部份電路組件的有效區位採用有有機半導體。 •採用單晶矽基板。 .採用玻璃基板。 •視訊的影格速率至少約爲25 Hz。 •只有真實空間影像資料會被傳輸給顯示器。 •影像資料中包含光強度及緩衝區對應資料。 •所進行的全像計算爲即時或近即時運算。 •所進行的全像計算將採用檢視表方法執行。 •將採用次全像影像來進行運算。 •顯示器在製造上將採用矽晶液晶技術。 •顯示器在製造上將採用MEMS技術。 120 200916986 •顯示器在製造上將採用場發射顯示器（FED)技術。 •使用高解析度顯示器的方法。 E.可透過合倂用於進行全像轉換和編碼的額外處理單元以一個擴充的圖形子系統的 3D著色管線(Rendering Pipeline)在同一基板上進行計算的全像影像顯示器 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行的全像顯示器，因此圖形子系統的3D著色管 線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額外處理單元。 •所進行的全像計算將使用位於顯示器的各個像素之間的電路組件來執行。 •全像計算會利用位於顯示器的像素矩陣以外的位置、但在與顯示器的像素所在的同 一基板上的電路組件來執行。 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器 的像素所在的同一基板上的電路組件來執行，且這些計算的本身並不牽涉到傅立葉轉換 或菲涅爾轉換的計算。 •計算會在分散在顯示器中個各個區域執行’以便針對各個分散區域，進行各個相應 的分散區域的像素編碼。 •電路組件中包括有薄膜電晶體。 •視訊的影格速率至少約爲25 Hz。 •只有真實空間影像資料會被傳輸給顯示器。 121 200916986 •影像資料中包含光強度及緩衝區對應資料。 •所進行的全像計算爲即時或近即時運算。 •所進行的全像計算將採用檢視表方法執行。 •採用次全像影像來進行運算。 •全像運算會均質的分散在整顯示器表面上。 •全像計算會分割爲拼貼在顯示器表面上的許多全等的微小部份（稱爲叢集）。 •根據上述構想，本案之全像顯示器是一個高解析度顯示器。 •一個虛擬觀察者視窗的直徑大約是眼睛瞳孔的直徑或更大的直徑。 •將可以顯示單色影像。 •將可以顯示彩色影像。 •在計算一個全像影像的一個像素的値時，只會考慮原始影像的一個次區段的値。 •用於進行全像影像重建的光並非在整個顯示器上完全一致，而是在顯示器的各個次 區段範圍內完全一致。 .全像轉換爲一維轉換。 .全像轉換爲二維轉換。 122 200916986 •備用電路組件（像是TFT)可能會被製造在像素矩陣的空間中，使這些電路組件可 以在發現某些用於裝置啓動的電路組件有故障時’用來取代某些在裝置啓動時會用到的 電路組件0 •將由物件所發出的波前將會在一個或多個虛擬觀察者視窗（VOW)中重建，且其中 —個三維景象（3D S)的每個單一物件點（OP)的重建只需要一個次全像影像 做爲要在SLM上進行編碼的整個全像影像（ΗΣ51_μ)的子集。 •在一個景象（3D S)離散化成爲多個物件點（OP)之後，針對3D景象的每個可見 的物件點，透鏡次全像影像（SHL)的複雜値會在SLM上進行編碼，其中透鏡次全像 影像的複雜値會利用下列方程式來決定：zL = exp { -i*〔（π/λί ) * ( x2 + y2 )〕}，其 中λ是參考波長，f是焦距，而x和y則是在次全像影像的平面中互相垂直的座標。 •棱鏡的次全像影像（SHP)會在全像影像平面（HE)中決定以便移動虛擬觀察者視 窗以遠離光軸。 •透鏡和稜鏡的是次全像影像是迴旋的，這也可以用SH = SHL * SHP來表示。 .每一個次全像影像（SH)會以一個統一分布的相位偏移來進行調制’這裡的相位偏 #每一個次全像影像都不盡相同。 •會將多個次全像影像疊加在一起來構成整個全像影像。 •用於進行重建的電腦產生全像影像的呈現可即時或近即時更新。 *在全像計算中會使用檢視表。 •各個物件點可以在用來進行重建的截頭錐體空間中的任何位置產生。 123 200916986 .用於第一顯示波長的z軸對應資料會進行兩次複製供第二及第三顯示波長使用。 •會針對三個顯示波長中的每一個顯示波長並行計算全像影像° •用於兩個色彩的色彩對應RGB內容會複製到個別的記憶體區段中’以確保三個分 色都可以獨立存取。 •用於每個顯示器色彩的透鏡函數和稜鏡函數會進行一項複乘法。 •顯示器的每一個叢集會套用一個隨機的相位。 •計算出來的SLM編碼會在全像顯示叢集中使用額外的運算法接受後續的處理。 •使用全像顯示器的方法。 F.可以以一個全像計算管線透過擴充顯示卡的3D管線(Pipeline)對三維空間中的各個 點進行循序全像轉換並同一基板上進行計算的全像影像顯示器 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行，使三維空間中的各個點的循序全像轉換藉由 透過一個全像計算管線來擴充顯示卡的3D管線(Pipeline)來執行的全像顯示器。 •所進行的全像計算將使用位於顯示器的各個像素之間的電路組件來執行。 •全像計算會利用位於像素矩陣以外的位置、但在與顯示器所在的同一基板上的電路 組件來執行。 •至少有一部分用來決定一個空間光麵器的編碼的計浦棚位於和空間光酬器 124 200916986 的像素所在的同一基板上的電路組件來執行，且這些計算的本身並不牽涉到傅立葉轉換 或菲涅爾轉換的計算。 •計算會在分散在顯示器中個各個區域執行，以便針對各個分散區域，進行各個相應 的分散區域的像素編碼。 •電路組件中包括有薄膜電晶體。 •視訊的影格速率至少約爲25 Hz。 •只有真實空間影像資料會被傳輸給顯示器。 •影像資料中包含光強度及緩衝區對應資料。 •所進行的全像計算爲即時或近即時運算。 •所進行的全像計算將採用檢視表方法執行。 •採用次全像影像來進行運算。 •全像運算會均質的分散在整顯示器表面上。 •全像計算會分割爲拼貼在顯示器表面上的許多全等的微小部份（稱爲叢集）。 •根據上述構想，本案之全像顯示器是一個高解析度顯示器。 •一個虛擬觀察者視窗的直徑大約是眼睛瞳孔的直徑或更大的直徑。 •將可以顯示單色影像。 125 200916986 •將可以顯示彩色影像。 •在計算一個全像影像的一個像素的値時，只會考慮原始影像的一個次區段的値。 •用於進行全像影像重建的光並非在整個顯示器上完全一致，而是在顯示器的各個次 區段範圍內完全一致。 •全像轉換爲一維轉換。 •全像轉換爲二維轉換。 •備用電路組件（像是TFT)可能會被製造在像素矩陣的空間中，使這些電路組件可 以在發現某些用於裝置啓動的電路組件有故障時，用來取代某些在裝置啓動時會用到的 電路組件。 •將由物件所發出的波前將會在一個或多個虛擬觀察者視窗（VOW)中重建’且其中 一個三維景象（3D S)的每個單一物件點（OP)的重建只需要一個次全像影像（SH) 做爲要在SLM上進行編碼的整個全像影像（tESLM)的子集。 •在一個景象（3D S)離散化成爲多個物件點（0P)之後，針對3D景象的每個可見 的物件點，透鏡次全像影像（SHL)的複雜値會在SLM上進行編碼，其中透鏡次全像 影像的複雜値會利用下列方程式來決定·· A = exp { -i*〔（πΜ ) * ( x2 + y2 )〕} ’其 中λ是參考波長，f是焦距，而x和y則是在次全像影像的平面中互相垂直的座標。 •稜鏡的次全像影像（SHP)會在全像影像平面（HE)中決定以便移動虛擬觀察者視 窗以遠離光軸。 •透鏡和稜鏡的是次全像影像是迴旋的，這也可以用SH = SHL * SHP來表示。 126 200916986 •每一個次全像影像（SH)會以一個統一分布的相位偏移來進行調制，這裡的相位偏 移每一個次全像影像都不盡相同。 •會將多個次全像影像疊加在一起來構成整個全像影像。 •用於進行重建的電腦產生全像影像的呈現可即時或近即時更新。 •在全像計算中會使用檢視表。 ♦各個物件點可以在用來進行重建的截頭錐體空間中的任何位置產生。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 •用於第一顯示波長的Z軸對應資料會進行兩次複製供第二及第三顯示波長使用。 .會針對三個顯示波長中的每一個顯示波長並行計算全像影像。 •用於兩個色彩的色彩對應RGB內容會複製到個別的記憶體區段中，以確保三個分 色都可以獨立存取。 •用於每個顯示器色彩的透鏡函數和稜鏡函數會進行一項複乘法。 •顯示器的每一個叢集會套用一個隨機的相位。 •計算出來的SLM編碼會在全像顯示叢集中使用額外的運算法接受後續的處理。 .全像計算可以在有完整的色彩對應及Z軸緩衝區資料可用之前開始。 127 200916986 •針對每一個次全像影像執行全像計算所需的時間小於一個影像框格的時間。 •針對每一個次全像影像執行全像計算所需的時間爲17 ms或以下。 •可以使用於軍事用途。 •顯示器的每一個叢集都有自己的檢視表用來儲存它所顯示的次全像影像的編碼。 •在讀取來自LUT的SH的內容之後，它會計算出目前顯示的SHCSHU)與新的SH (SHn)之間的差。 •三維空間中的各個點的循序全像轉換，會透過擴充一個具全像計算管線(Pipeline)之 顯示卡之3D管線的方法來執行，並不限於特定種類的SLM。 •使用全像顯示器的方法。 G.可在同一基板上執行計算的全像影像顯不器，全像顯示器並可隨機編址 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行，使連續的真實空間影像框格之間用在全像計 算中的真實空間影像資料各不相同’而全像顯示資料會以次全像影像差異資料和顯示記 憶位置資料的形式傳送給全像顯示叢集的全像顯示器。 •二維空間中的各個點的循序全像轉換會透過擴充一個具全像計算管線(Pipeline)之顯 示卡之3D管線的方法來執行。 .所進行的全像計算將使用位於顯示器的各個像素之間的電路組件來執行。 128 200916986 •全像計算會利用位於像素矩陣以外的位置、但在與顯示器所在的同一基板上的電路 組件來執行。 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器 的像素所在的同一基板上的電路組件來執行，且這些計算的本身並不牽涉到傅立葉轉換 或菲涅爾轉換的計算。 •計算會在分散在顯示器中個各個區域執行，以便針對各個分散區域，進行各個相應 的分散區域的像素編碼。 •電路組件中包括有薄膜電晶體。 •視訊的影格速率至少約爲25 Hz。 .只有真實空間影像資料會被傳輸給顯示器。 •影像資料中包含光強度及緩衝區對應資料。 •所進行的全像計算爲即時或近即時運算。 •所進行的全像計算將採用檢視表方法執行。 •會顯示各個次全像影像。 •全像運算會均質的分散在整顯示器表面上。 •全像計算會分割爲拼貼在顯示器表面上的許多全等的微小部份（稱爲叢集）。 •根據上述構想，本案之全像顯示器是一個高解析度顯示器。 129 200916986 •一個虛擬觀察者視窗的直徑大約是眼睛瞳孔的直徑或更大的直徑。 .將可以顯示單色影像。 •將可以顯示彩色影像。 .在計算一個全像影像的一個像素的値時’只會考慮原始影像的一個次區段的値。 •用於進行全像影像重建的光並非在整個顯示器上完全一致’而是在顯示器的各個次 區段範圍內完全一致。 •全像轉換胃一維轉換。 •全像轉換爲二維轉換。 •備用電路組件（像是TFT)可能會被製造在像素矩陣的空間中，使這些電路組件可 以在發現某些用於裝置啓動的電路組件有故障時’用來取代某些在裝置啓動時會用到的 電路組件。 •將由物件所發出的波前將會在一個或多個虛擬觀察者視窗（vow)中重建，且其中 一個三維景象（3D S)的每個單一物件點（OP)的重建只需要一個次全像影像（SH) 做爲要在SLM上進行編碼的整個全像影像（fESLM)的子集。 •在一個景象（3D S)離散化成爲多個物件點（OP)之後，針對3D景象的每個可見 的物件點，透鏡次全像影像（SHL)的複雜値會在SLM上進行編碼’其中透鏡次全像 影像的複雜値會利用下列方程式來決定：zL = exp { -i*〔（π/λί ) * ( x2 + y2 )〕丨’其 中λ是參考波長，f是焦距，而χ和y則是在次全像影像的平面中互相垂直的座標。 130 200916986 •稜鏡的次全像影像（SHP)會在全像影像平面（HE)中決定以便移動虛擬觀察者視 窗以遠離光軸。 •透鏡和稜鏡的是次全像影像是迴旋的，這也可以用SH = SHL * SHP來表示。 .每一個次全像影像（SH)會以一個統一分布的相位偏移來進行調制，這裡的相位偏 移每一個次全像影像都不盡相同。 •會將多個次全像影像疊加在一起來構成整個全像影像。 .用於進行重建的電腦產生全像影像的呈現可即時或近即時更新。 •各個物件點可以在用來進行重建的截頭錐體空間中的任何位置產生。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 •用於第一顯示波長的Z軸對應資料會進行兩次複製供第二及第三顯示波長使用。 •會針對三個顯示波長中的每一個顯示波長並行計算全像影像。 •用於兩個色彩的色彩對應RGB內容會複製到個別的記憶體區段中，以確保三個分 色都可以獨立存取。 •用於每個顯示器色彩的透鏡函數和稜鏡函數會進行一項複乘法。 •顯示器的每一個叢集會套用一個隨機的相位。 •計算出來的SLM編碼會在全像顯示叢集中使用額外的運算法接受後續的處理。 131 200916986 •可以使用於軍事用途。 •各個全像計算單元會接收影像差異資料。 •如果一個特定叢集的連續框格的顯示資料之間沒有差異、或者只有微乎其微而可以 忽略的差異，則並不須要傳送資料給叢集。 •每一個全像轉換單元都會傳送相對於重建點或用來在SLM上進行編碼的點的3D 差異點影像資料。 •在每一個全像顯示叢集中有一個分解器，它會將計算出來的全像影像顯示資料分解 爲次全像影像資料以及大小和位置資訊，其中後面這兩個値可以用來計算次&amp;胃§彡胃# RAM中的位址範圍，使次全像影像SH或SHD的資料可以被寫入到叢集中的正確 SLM像素格上。 •使用一個特殊的隨機存取記憶體（RAM)來使輸入側只會寫入新的SH或SHD，而 在輸出側則會完整的讀取整個記憶體並將完整的資訊寫入到SLM上。 •使用全像顯示器的方法。 H.在像素空間中具備有運算功能的顯示器 運算功能會由佈置在與顯示器的像素所在的同一基板上的電路組件執行的顯示器。 •運算功能會由佈置在與顯示器的像素之間的電路組件執行。 .運算功能會由佈置在像素矩陣以外、但在與顯示器所在的同一基板上的電路組件執 行0 132 200916986 •顯示器上的顯示資料的延遲小於由佈置在與顯示器的像素所在的同一基板上的電路 組件所執行的運算功能如果在其它地方執行所造成的的延遲。 •這些運算是圖形的運算。 •可以構成高速遊戲裝置的一部份。 •可以使用於軍事用途。 •計算會在分散在顯示器中個各個區域執行，以便針對各個分散區域’進行各個相應 的分散區域的像素編碼。 •電路組件中包括有薄膜電晶體。 •至少有一部份電路組件的有效區位採用有多晶矽。 •至少有一部份電路組件的有效區位採用有連續晶粒矽。 •至少有一部份電路組件的有效區位採用有多晶矽鍺。 •至少有一部份電路組件的有效區位採用有單晶矽。 •影像資料的影格速率至少約爲25 Hz。 •運算（可能是並行運算）會分割爲拼貼在顯示器表面上的許多全等的微小部份，稱 •顯示器可以透過將許多全等的叢集拼貼在一起而構成。 133 200916986 •顯示器是一個筒解析度顯市器。 •顯示器是一個極高解析度顯示器。 .將可以顯示彩色影像。 •將可以顯示RGB格式的彩色影像。 •顯示器在製造上將採用矽晶液晶技術。 •有一個額外的邏輯可用於區域轉送現有計算所得的資料，而這個額外的邏輯也同時 用於轉送原始影像給各個叢集’所以至少可以剔除某些共通的橫列和縱列線路。 •使用顯示器的方法。 I.吸收 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行，並且可以確保距離虛擬觀察者視窗較近的物 件點會遮蔽沿著相同的視線上距離虛擬觀察者視窗較遠物件點的全像顯示器。 •計算的本身並不牽涉到傅立葉轉換或菲淫爾轉換的計算。 •全像影像編碼資料會在像素矩陣所佔用的空間以外的地方計算，然後再將這些全像 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 134 200916986 外處理單元。 •三維空間中的各個點的循序全像轉換會透過使用一個全像計算管線來擴充顯示卡的 3D管線(Pipeline)來執行。 •用於進行全像計算中的真實空間影像資料是每一格連續真實空間影像框格之間的 差，而全像顯示資料會以次全像影像差異資料以及顯示記憶位置資料的形式傳送給各個 全像顯示叢集。 •“吸收”會利用由佈置在與像素矩陣所在的同一基板上的電路組件所執行的計算來實 施。 •“吸收”會利用由佈置在顯示器的像素之間的電路組件所執行的計算來實施。 •一個虛擬觀察者視窗的直徑大約是眼睛瞳孔的直徑或更大的直徑。 • VOW會分成兩個或多個區段。 •每一個vow區段的大小大致和人類眼睛的瞳孔大小相當。 •每一個vow區段會以一個不同的次全像影像進行編碼。 • “吸收”會在構成緩衝區對應資料和光強度對應資料的階段中進行。 •使用全像顯示器的方法。 J.顯示卡的功能 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 135 200916986 像麵在的同—麵上的電腦件雜行，且顯示卡的功能會棚讎在麵示器的像 素所在的同一·基板上的電路組件來執行的全像顯示器。 •計算的本身並不牽涉到傅立葉轉換或菲涅爾轉換的計算° •全像影像編碼資料會在像素矩陣所佔用的空問以外的地方計算’然後再將這些全像 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 .圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 •三維空間中的各個點的循序全像轉換會透過使用一個全像計算管線來擴充顯示卡的 3D管線(Pipeline)來執行。 •用於進行全像計算中的真實空間影像資料是每一格連續真實空間影像框格之間的 差，而全像顯示資料會以次全像影像差異資料以及顯示記憶位置資料的形式傳送給各個 全像顯示叢集。 •顯示卡的功能會利用佈置在顯示器的像素之間的電路組件來執行。 •顯示卡的功能會利用佈置在像素矩陣以外的電路組件來執行。 •顯示卡的功能包括紋理映射。 .顯示卡的功能包括呈現多邊形。 •顯示卡的功能包括將頂點轉譯成不同的座標系統。 136 200916986 •顯示卡的功能包括可程式著色引擎。 •顯示卡的功能包括超取樣和內插技術來減少疊影。 •顯示卡的功能包括極高精密度的色彩空間。 •顯示卡的功能包括2D加速計算能力。 • .顯示卡的功能包括影格緩衝區能力。 •顯示卡的功能包括“動畫專家小組”（MPEG)基元。 •顯示卡的功能包括執行牽涉到矩陣和向量操作的運算。 •顯示卡的功能包括採用一個由佈置在與像素矩陣所在的同一基板上的TFT所執行的 3D 著色管線(Rendering Pipeline)。 •使用全像顯示器的方法。 K. 2D㈠3D轉換 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行，並且會執行2D ^ 3D影像轉換的全像顯 示器。 •計算的本身並不牽涉到傅立葉轉換或菲涅爾轉換的計算。 •全像影像編碼資料會在像素矩陣所佔用的空間以外的地方計算，然後再將這些全像 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 137 200916986 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 .三維空間中的各個點的循序全像轉換會透過使用—個全像計算管線來擴充顯示卡的 3D管線(Pipeline)來執行。 •用於進行全像計算中的真實空間影像資料是每一格連續真實空間影像框格之間的 差，而全像顯示資料會以次全像影像差異資料以及顯示記憶位置資料的形式傳送給各個 全像顯示叢集。 .2D㈠3D影像轉換會利用佈置在與顯示器的像素所在的同一基板上的電路組件來 執行。 • 2D㈠3D影像轉換會利用不在與顯示器的像素所在的同一基板上的電路組件來執 行。 • 2D㈠3D影像轉換會利用佈置在顯示器的像素之間的電路組件來執行。 • 2D㈠3D影像轉換會利用在像素矩陣以外但在與像素所在的同一基板上的電路組 件來執行。 • 2D β 3D影像轉換會利用成對的立體影像來執行。 •顯示裝置會根據所接收到的資料計算出一個具有相應的緩衝區對應資料的二維（2D ) 影像。 •用來執行2D㈠3D轉換的電路組件有權存取一個包含有一組已知的3D形狀的資 138 200916986 料庫。 •用來執行2D e 3D $專換的電路組件有權存取一個包含有一組已知的2D外形的資 料庫’在這個資料庫中它可能會嘗試匹配輸入的2D影像資料。 • 2D 〇 3D影像轉換會根據一個單獨的、非自動立體照相的2D影像來執行。 •使用全像顯示器的方法。 L.視訊會談（3DSkype™) 全像顯示器將可以提供網際網路語音及全像影像協定（VHIOIP)服務。 •至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器 的像素所在的同一基板上的電路組件來執行。 •計算的本身並不牽涉到傅立葉轉換或菲涅爾轉換的計算。 •全像影像編碼資料會在像素矩陣所佔用的空間以外的地方計算，然後再將這些全像 ' 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 •三維空間中的各個點的循序全像轉換會透過使用一個全像計算管線來擴充顯示卡的 3D管線(Pipeline)來執行。 •用於進行全像計算中的真實空間影像資料是每一格連續真實空間影像框格之間的 139 200916986 差，而全像顯示資料會以次全像影像差異資料以及顯示記憶、位置資料的形式傳送給各 全像顯示叢集。 ι、口 δ •可以提供VHIOIP點對點通訊服務。 •可以提供檔案分享。 .可透過它所連線的一個全球網路提供即時訊息服務。 •可透過它所連線的一個電腦網路提供通訊服務。 •可透過它所連線的一個電腦網路提供檔案分享服務。 •可透過它所連線的一個電腦網路提供即時訊息服務。 •提供有可供線上暫時使用、不可下載的電腦軟體以允許用戶使用VHIOIP通訊服務。 .提供有可供下載的線上軟體以允許用戶使用VHIOIP通訊服務。 .提供有可進入網域及網域資料庫系統以存取全像顯示資料的通道。 •使用全像顯示器的方法。 M.編碼補償 可以在編碼步驟中或編碼步驟之前對全像影像資料施加補償，以提供一個可以看得更清 楚的影像的全像顯示裝置。 .至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器 140 200916986 的像素所在的同一基板上的電路組件來執行。 •計算的本身並不牽涉到傅立葉轉換或菲淫爾轉換的計算。 •全像影像編碼資料會在像素矩陣所佔用的空間以外的地方計算，然後再將這些全像 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 •三維空間中的各個點的循序全像轉換會透過使用一個全像計算管線來擴充顯示卡的 3D管線(Pipeline)來執行。 •用於進行全像計算中的真實空間影像資料是每一格連續真實空間影像框格之間的 差，而全像顯示資料會以次全像影像差異資料以及顯示記憶位置資料的形式傳送給各個 全像顯示叢集。 •“補償”會利用佈置在與顯示器的像素所在的同一基板上的電路組件來實施。 .“補償”會利用佈置在顯示器的各個像素之間的電路組件來實施。 • “補償”會在編碼步驟中施加到全像影像資料上。 •“補償”會在編碼步驟之前施加到全像影像資料上。 •會對一個以明亮的調子爲主而傾向於可能會曝光不足的景象施加補償來進行修正。 •會對一個以陰暗的調子爲主而傾向於可能會曝光過度的景象施加補償來進行修正。 141 200916986 •使用全像顯不器的方法。 N.眼睛追蹤 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行，並且會實施眼睛追蹤的全像顯示器。 •計算的本身並不牽涉到傅立葉轉換或菲涅爾轉換的計算。 •全像影像編碼資料會在像素矩陣所佔用的空間以外的地方計算，然後再將這些全像 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 ” ’ .三維空間中的各個點的循序全像轉換會透過使用一個全像計算管線來擴充 3D管線(Pipeline)來執行。 一 下的 •用於進行全像計算中的真實空間影像資料是每一格連續颠空間影像 差’而全麵不資料會以次全像影像差異資料以及顯示記雖黯料的形 間的 全像顯示叢集。 各個 • π]以針對單一觀看者實施眼睛追％。 可以針對複數觀看者實施眼晴追_。 眼睛追蹤會邏麵麵者的_來限定搜尋棚， 然後透過偵測眼睛來• LC0S • Digital Light Processing (DLP) technology The list of possible display technologies that can be used in one display design of the present invention is as follows: Liquid Crystal Display (LCD) - Category LC0S 矽 Liquid Crystal 96 200916986 NLC Nematic Liquid Crystal TN Twisted Nematic VAN Vertical Arranged nematic FLC Strongly induced liquid crystal FED (field emission display) SED surface conduction electron emission display Nano barrier tube (4) (based on miscellaneous or coated oxidation (based on the substrate), but these can only be used As a light source, because it emits non-uniform light) Motor System Mirror Array / Digital Light Processing (DLP) Technology MEMS Mirror (Micro Motor System) 'also known as M〇EMS (Micro Photoelectric System) Full Image The methods of image calculation include: - Viewing Table (LUT) - Analytical Operation - Method described in Patent Publication No. WO 2006/066919, which is incorporated herein by reference in its entirety. - ray tracing method 97 200916986 Conversion category: -2D conversion - 1D conversion in horizontal plane - 1D conversion in vertical plane Coding categories: -Burckhardt coding - phase-only coding - bi-phase coding - BIAS coding - 瞧 (minimum Off-coded—An external external holographic image calculation unit that uses more than three pixels of each hologram pixel, which may include – for high-order intestines or containing approximately 520 million transistors and · MHz The line frequency is based on the application of the integrated circuit (ASIC) or a pure custom integrated circuit (10). To transmit about 23 〇 to the low voltage differential signal (10) s) to the display, each pair of transmissions You can use the sum of i(3) per second. In order to receive data, a glass flip-chip (C〇G) course and a tandem line drive device are also required. If the operation is to be integrated on the display, then the high-frequency source material visual interface (Wi) receiver must be placed on the extra hardware. The original data must only be transmitted at a rate of 5 times lower than the transmission rate of 200916986 (see Figure D. It is possible to use a low-bay electronic system with only a few connections to the display. This electronic system probably Similar to today's electronic systems in low-resolution 2D TFT displays. Note: U Section II outlines possible manufacturing methods without departing from the scope of the invention.  Features. [In the drawings of this document, the relevant dimensions shown are not necessarily to scale. Various modifications and alterations of the present invention will be apparent to those skilled in the art and the scope of the present invention will not be exceeded. Those skilled in the art should understand that the present invention is not Limited to the illustrated examples attached to this document. The j file contains a variety of concepts (as described in "Concept A - T"). The appendix contains content that can help define these three concepts. Those familiar with this type of technology will be very clear, and the idea of a concept may help _ other silk. Certain ages in this document may constitute v + indispensable - part ' and will be clearly stated elsewhere in this document. 99 200916986 Appendix.  Introduction to Technology The following section will be used as a basic introduction to a number of key techniques used in certain systems in which the present invention is applied. In traditional holographic techniques, the observer can see a holographic image reconstruction of an object (this may be a constantly changing scene); however, the distance between the observer and the holographic image may not be relevant. This reconstruction, in the typical visual configuration, is on or near the image plane of the light source that illuminates the holographic image, that is, on or near the Fourier plane of the holographic image. Therefore, this reconstructed image has the same far-field light distribution as the reconstructed real-world object. An early system (as described in the patents of WO 2 〇 04/044 659 and US 2006/0055994, the entire disclosure of which is incorporated herein by reference in its entirety in the entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire entire all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all all On the Fourier plane of the hologram or near it, on the contrary, the shell IJ is a virtual observer window area on the Fourier plane of the hologram; the observer places the eye at this position and will only see A correctly reconstructed image. The holographic image is encoded on an LCD (or other type of spatial light modulator) and illuminated by an optical design so that the Fourier transform of the holographic image is performed on the virtual observer window (hence a Fourier transform) The mapping is mapped to the eye; after the reconstructed object formed in a truncated cone space between the observer window and the SLM, the Fresnel transformation of the holographic image is better described in terms of propagation because it Not on the focal plane of the lens. Instead, it is defined by a near-field light distribution (using a spherical wavefront configuration, as opposed to a plane wavefront with a far-field distribution). This reconstruction can occur anywhere in the virtual observer window (as described above, the Fourier plane of the holographic image) and the SLM or even appear as a virtual object behind the SLM. This approach has many consequences. First, the basic limitation for holographic video system designers is the pixel pitch problem of SLM (or other types of light modulators). The goal is to achieve large holographic image reconstruction using SLMs that are commercially available at reasonable cost and with appropriate pixel spacing. But in the past, this was not possible due to the following factors. The periodic interval between adjacent diffraction orders in the Fourier plane is represented by λ/)/;?, where λ is the wavelength of the light of 100 200916986, which is the distance from the holographic image to the Fourier plane; Is the pixel pitch of the SLM. But on a traditional hologram display, the reconstructed object appears on the Fourier plane or its attachment. Therefore, a reconstructed object must remain less than the periodic interval; if larger, its edges will be obscured by the reconstruction of an adjacent diffraction order. This will make the reconstructed object very small, usually only a few cm wide, even if it is a dedicated small-pitch display that is costly. However, with the method of the present invention, the size of the virtual observer window (e.g., the number of locks above, which is located on the Fourier plane of the hologram) is only as large as the pupil of the eye. The result of this is that even SLMs of medium pitch size can be used. At the same time, because the reconstructed object can completely fill the frusto-cone space between the virtual observer window and the omni-image, it is indeed large enough, and certainly much larger than the periodic interval. In addition, if an OASLM is used, the Bay IJ will not have a jagged phenomenon and therefore a periodic problem, so there will be no more restrictions that the virtual observer window must remain less than one cycle interval. There is another benefit to adopting another variant. When calculating a holographic image, let's start with our understanding of the reconstructed object. For example, you might have a 3D image file of a racing car. This file will show you which different viewing positions the object should be visible from. In traditional holographic techniques, omni-image must produce a reconstruction of a car directly from a 3D image archive in a process that requires intensive operations. But the virtual observer window approach can take a different and more efficient technique. Starting from a plane of the reconstructed object, we can calculate the virtual observer window by means of the Fresnel transformation of the object. We can then perform this action on all object planes and sum up all the results to produce a cumulative Fresnel transformation; this will define the wavefield of the entire virtual observer window. We then calculate the hologram image by means of the Fourier transform of this virtual observer window. Since the virtual observer window contains all the information about the object, only a single planar virtual observer window (rather than a multi-plane object) must be Fourier transformed into a holographic image. This is particularly advantageous if the transition from the virtual observer window to the holographic image is not a single conversion step but a repeated conversion step (such as a repeated Fourier transform algorithm). If an iterative operation is required, each iteration step includes only the Fourier transform of one virtual observer window rather than the Fourier transform of each object plane, which significantly reduces the computational requirements. Another beneficial result of the virtual observer window approach is that all the information needed to reconstruct a particular object point is contained in a relatively small portion of the hologram image; this is different from traditional holographic techniques (it reconstructs The information required for a particular 101 200916986 object point is scattered throughout the hologram image. Since we only need to encode this information into a significantly smaller portion of the hologram, this means that we need to process and encode much less information than a traditional hologram. This also means that even with instant video holography, traditional computing devices can be used (for example, a traditional DSP with cost and performance suitable for mass production of marketed devices). However, it also has some results that may not be satisfactory. First, the viewing distance relative to the holographic image will be very important. A holographic image is encoded and projected in such a way that the correct reconstruction can only be seen when the eye is in or near the Fourier plane of the holographic image; In general holographic images, viewing distance is not an important factor. However, a variety of different techniques have been used to reduce the sensitivity of the Z-axis distance or the design difficulty surrounding this factor. At the same time, since the holographic image is reconstructed and illuminated in such a way that the correct holographic image reconstruction can only be from a precise and minimal viewing position (especially for lateral positioning and Z-axis distance) The way you see it is encoded and projected, so eye tracking may be required. Due to the extreme sensitivity to the Z-axis distance, various techniques are needed to reduce the sensitivity of the X, Y orientation or the design difficulty surrounding this factor. For example, as the pixel pitch shrinks (because it has the manufacturing advantage of SLM), the size of the virtual viewer window can be increased. In addition, more efficient coding techniques (such as Kinoform coding) also facilitate the use of a larger portion of the periodic interval as a virtual observer window, and thus can increase the virtual observer window. One of the premise of the above description is to assume that we are facing a Fourier hologram image. The face viewer window is on the Fourier plane of the hologram image, that is, on the image plane of the light source. It has the advantage that non-diffracted light will focus on the so-called DC point. If the virtual face is not in the image of the light source, this can be used on the Fresnel hologram. However, it must be noted that due to a disturbing background, non-diffracted light is invisible. It must be noted that Peng-Compense Steel's "conversion" of Lin Guanbao is a mathematical or computational technique that is quite similar to the transformation of the singer. The Maxwellian wave equation has a more precise definition of the transformation that only approximates the physical process; the Philippine face is a second-order approximation, but it has the following 雠: (a) because they are opposite to the differential Algebra, so it can be handled in a more computationally efficient way 'and (ii) can be accurately implemented in an optical system. 102 200916986 Further details can be found in U.S. Patent Application Nos. US 2006-0138711, US 2006-0139710, and US 2006-0250671, the entire contents of each of which are incorporated herein by reference. 200916986 mm ii Description of terms used in this document Computer produces a holographic image A computer-generated hologram image (CGH) is a holographic image calculated from a scene. The CGH may include a composite parameter 代表 representing the amplitude and phase of the light waves needed to reconstruct the scene. CGH can be calculated, for example, by "consistent ray tracing", by interference between a simulated scene and a reference light wave, or by Fourier or Fresnel conversion. Encoding An encoding is a program in which control of a holographic image is provided to a spatial light modulator (for example, a pixel grid that makes up it, or a contiguous region of a continuous SLM such as OASLM). In general, a holographic image includes a composite number 値 representing amplitude and phase. Encoding area The coding area is usually a space-limited area in a holographic image in which holographic image information for a single scene point is encoded. Spatial constraints can be achieved by a sudden truncation or by a smooth truncation through a Fourier transform from the virtual observer window to the holographic image. Fourier transform Fourier transform is used to calculate the propagation of light in the far field of a spatial light modulator. The wavefront is described by a plane wave. Fourier Plane The Fourier plane contains the Fourier transform of the distribution of light over a spatial light modulator. Without any focal lens, the Fourier plane would be in infinity. If there is a focal lens in the path of travel of the light close to the spatial light modulator, the Fourier plane corresponds to the plane containing the image of the source. Fresnel conversion Fresnel conversion is used to calculate the propagation of light in the near field of a spatial light modulator. The wave front is described by a spherical wave. The phase factor of the light wave includes a quadratic term that depends on the square of the lateral coordinate. 104 200916986 Frustum The virtual observer window and the SLM form a virtual frustum and extend all the way to the rear of the SLM. The scene will be reconstructed in this frustocone. The size of the reconstructed scene is limited by this frustum and is not limited by the periodic interval of the SLM. Light source system The light source system may include a uniform light source (such as a laser) or a partially consistent light source (such as an LED). Consistent temporal and spatial consistency of the partially consistent light source must be sufficient to facilitate good scene reconstruction' emission tables.  The spectral line width and lateral extent must be sufficient /],. Virtual Observer Window (vow) ‘Virtual Observer Window is a virtual window on the viewer's plane. Through this you can see the reconstructed 3D object. Vow is a Fourier transform of a holographic image and is located within a periodic interval to avoid seeing multiple reconstructed images of the object. The size of the vow must be at least the size of the pupil of the eye. If at least one VOW is placed at the observer's eye and has an observer tracking system, the vow is smaller than the observer's lateral movement. This makes it easy to use an SLM that only has the appropriate resolution (and therefore also a small periodic interval). Think of VOW as a keyhole through which we can see reconstructed 3D objects; either one VOW per eye or one VOW for both eyes. Period Interval If the CGH is displayed on an SLM consisting of individual addressable pixels, it is sampled. This sampling action produces a periodic repetition of the diffractive pattern. The period interval is expressed in AD/p, where λ is the wavelength, D is the distance from the hologram to the Fourier plane, and ρ is the pitch of the SLM pixel. However, OASLM does not sample, so there is no periodic repetition of the diffraction pattern; this repetition is in fact suppressed. Reconstruction Illuminating a spatial light modulator with omni-image encoding will reconstruct the original distribution of light. This distribution of light is used to calculate the hologram image. In theory, the observer will not be able to distinguish the reconstructed light distribution from the original light distribution. On most of the holographic displays, what is reconstructed is the distribution of the light of the scene; on our display, the distribution of the light in the window of the virtual observer 105 200916986 is reconstructed. The scene to be reconstructed is a real or computer-generated three-dimensional light distribution. In a special case, it could also be a two-dimensional light distribution. A scene may include various fixed or moving objects arranged in a space. Spatial Light Modulator (SLM) The SLM is used to modulate the wavefront of the incoming light. An ideal SLM will be able to represent an arbitrary composite number 値, and also a good P, which can control the amplitude and phase of an input light wave, respectively. However, a typical SLM will only control one property, not amplitude or phase, and there may be adverse effects that may also affect another property. 106 200916986 Appendix Minor This document contains a variety of commemorations (as described in "Music A~T"). The following instructions may help define these commemorations. A. At least a portion of the holographic image display that can be computed on the same substrate as the pixel is used to determine that the calculation of the encoding of a spatial light modulator utilizes circuit components located on the same substrate as the pixel of the spatial light modulator. A full-image display that is executed. • At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located between the individual pixels of the spatial light modulator. . The calculations are performed in a plurality of regions dispersed in the display to perform pixel coding of respective corresponding dispersed regions for each of the dispersed regions. • A thin film transistor is included in the circuit assembly. . At least some of the circuit components have polysilicon in their effective locations. • At least some of the circuit components have active areas with continuous die. • At least some of the circuit components have polycrystalline mortars in their effective locations. • At least some of the circuit components have a single crystal 有效 in the effective location. • At least some of the circuit components have a valid area with a single die. 107 200916986 • At least some of the circuit components have an effective location with organic semiconductors. • Use a single crystal germanium substrate. . A glass substrate is used. . Only real-world image data will be transmitted to the display. • Video frame rate is at least approximately 25 Hz. • The image data contains light intensity and buffer corresponding data. • The holographic calculations performed are either immediate or near real-time operations. • The holographic calculations performed will be performed using the view method. • Use sub-holographic images for calculations. • Data used to add sub-holographic images is exchanged over a sub-image size. • The holographic operation is evenly distributed across the entire display surface. • The holographic calculation is divided into many congruent tiny parts (called clusters) that are tiled on the surface of the display. • Data used to join sub-holographic images is exchanged at a cluster scale distance. • A holographic display can be constructed by tiling a number of congruent clusters together. The . According to the above concept, the holographic display of the present case is a high-resolution display. • According to the above concept, the holographic display of this case is an extremely high resolution display. • The diameter of a virtual observer window is approximately the diameter of the pupil of the eye or a larger diameter. • This virtual observer window has a diameter of approximately 1 cm or more. • A pair of buffer correspondences and light intensity correspondence data are organized for one parent (ie, for each virtual observer window). • A monochrome image will be displayed. • The color image will be displayed. • Color images in RGB format will be displayed. • When calculating the 値 of a pixel of a holographic image, only the 値 of a sub-segment of the original image is considered. % • The light used for holographic image reconstruction is not exactly the same across the display, but is exactly the same across the sub-sections of the display. . It is only necessary to transfer less than the transmission of holographic image data to transfer the original image data. • Reducing the data transmission frequency has the benefit of reducing the power consumption in the row and column drives. • Most of the pixel area required for the tandem and horizontal lines in previous patented solutions can be used for other purposes. 109 200916986 • The area of the transparent electrode can be increased, and thus the transmittance of the display can be improved. • The display panel can be controlled using traditional display technology. • The display will be manufactured using twin crystal technology. • The display will be manufactured using MEMS technology. • The display will be manufactured using Field Emission Display (FED) technology. • The hologram is converted to a one-dimensional conversion. • The hologram is converted to a two-dimensional transform. • There is an additional logic for the area to transfer the existing calculated data, and this additional logic is also used to forward the original image to each cluster, so at least some common course and column lines can be eliminated. . Alternate circuit components (such as TFTs) may be fabricated in the space of the pixel matrix, so that these circuit components can be used to replace some of the device startups when it is found that some of the circuit components used for device startup are faulty. The circuit components to. • The method of using a hologram display. B. At least a portion of the holographic image display that can be computed on the same substrate for efficient spatial light modulator encoding calculations is used to determine the calculation of the encoding of a spatial light modulator using the pixels located in the spatial light modulator The hologram display is implemented by circuit components on the same substrate, and the calculations themselves do not involve the calculation of Fourier transform or Fresnel transform. 110 200916986 • At least a portion of the computation used to determine the encoding of a spatial light modulator is performed using circuit components located between the individual pixels of the spatial light modulator. • Calculations are performed in various areas scattered throughout the display to enable pixel encoding of the respective discrete areas for each discrete area. • A thin film transistor is included in the circuit assembly. • At least some of the circuit components have polysilicon in their active locations. . At least some of the circuit components have active regions with continuous die. • At least some of the circuit components have polysilicon in their active locations. . At least some of the circuit components have a single crystal 有效 in the effective location. • At least some of the circuit components have a valid area with a single die. • At least some of the circuit components have active semiconductors in their active locations. • Use a single crystal germanium substrate. . A glass substrate is used. . Only real-world image data will be transmitted to the display. • Video frame rate is at least approximately 25 Hz. 111 200916986 • The image data contains the light intensity and buffer corresponding data. • The holographic calculations performed are either immediate or near real-time operations. • The holographic calculations performed will be performed using the view method. • Use sub-holographic images for calculations. • The holographic operation is evenly distributed across the entire display surface. • The holographic calculation is divided into many congruent tiny parts (called clusters) that are tiled on the surface of the display. • According to the above concept, the holographic display of this case is a high-resolution display. • The diameter of a virtual observer window is approximately the diameter of the pupil of the eye or a larger diameter. • A monochrome image will be displayed. • The color image will be displayed. • When calculating the 値 of a pixel of a holographic image, only the 値 of a sub-segment of the original image is considered. • The light used to reconstruct the holographic image is not exactly the same across the display, but is exactly the same across the sub-sections of the display. • Only enough lines are needed to transfer the original image data than to transfer the holographic image data. • Reducing the data transmission frequency has the benefit of reducing the power consumption in the row and column drives. 112 200916986 • Most of the pixel area required for the tandem and horizontal lines in previous patented solutions can be used for other purposes. • The area of the transparent electrode can be increased, and thus the transmittance of the display can be improved. • The display panel can be controlled using traditional display technology. • The display will be manufactured using twin crystal technology. • The display will be manufactured using MEMS technology. • The display will be manufactured using Field Emission Display (FED) technology. • The hologram is converted to a one-dimensional conversion. . The hologram is converted to a two-dimensional transform. • There is an additional logic for the area to transfer the existing calculated data, and this additional logic is also used to forward the original image to each cluster, so at least some common course and column lines can be eliminated. • Alternate circuit components (such as TFTs) may be fabricated in the space of the pixel matrix so that these circuit components can be used to replace some of the device startups when it is found to be faulty. The circuit components used. • The wavefront emitted by the object will be reconstructed in one or more virtual observer windows (VOW), and the reconstruction of each single object point (OP) of one of the three-dimensional scenes (3D S) requires only one full The image (SH) is a subset of the entire hologram (PESLM) to be encoded on the SLM. • After a scene (3D S) is discretized into multiple object points (OP), for each object point of the 3D scene visible 113 200916986, the complex sub-image of the lens (SHL) will be encoded on the SLM. The complex 値 of the lens sub-image is determined by the following equation: ZL = eXp { -i*[(W ) * ( X2 + y2 Sichuan, where λ is the reference wavelength 'f is the focal length' and X and y The coordinates of the bank perpendicular to each other in the plane of the sub-image. • The sub-holographic image (SHP) of the bank is determined in the holographic image plane (HE) to move the virtual observer window away from the optical axis. The hologram image is convoluted, which can also be represented by SH = SHL * SHP. • Each sub-image (SH) is modulated with a uniformly distributed phase offset, where the phase The offset is different for each sub-image. • Multiple sub-images are superimposed to form the entire holographic image. • The computer used for reconstruction produces a holographic image that can be instantly or nearly instantaneously rendered. Update. • Check in hologram calculation Tables • Individual object points can be generated at any position in the frustoconic space used for reconstruction • Use of a holographic display method C-Full image display that can be decompressed on the same substrate The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then the holographic image encoding data is compressed by a known data compression technique and then transmitted to circuit components on the display substrate. The holographic display of the decompressed function is then performed on the received data. • At least a portion of the calculation used to determine the encoding of a spatial light modulator utilizes the same substrate as the pixel of the spatial light modulator 114 200916986. The circuit components are implemented. • The circuit assembly includes a thin film transistor. At least some of the circuit components have polysilicon in their effective locations. • At least some of the circuit components have active areas with continuous die. • At least some of the circuit components have polysilicon in their active locations. • At least some of the circuit components have a single crystal 有效 in the effective location. • At least some of the circuit components have a valid area with a single die. • At least some of the circuit components have active semiconductors in their active locations. . A single crystal germanium substrate is used. . A glass substrate is used. • Video frame rate is at least approximately 25 Hz. • The image data contains light intensity and buffer corresponding data. • The holographic calculations performed are either immediate or near real-time operations. • The holographic calculations performed will be performed using the view method. . The sub-holographic image is used for the calculation. 115 200916986 • According to the above concept, the holographic display of this case is a high-resolution display. • The diameter of a virtual observer window is approximately the diameter of the pupil of the eye or a larger diameter. • A monochrome image will be displayed. • The color image will be displayed. • When calculating the 値 of a pixel of a holographic image, only the 値 of a sub-segment of the original image is considered. • The light used to reconstruct the holographic image is not exactly the same across the display, but is exactly the same across the sub-sections of the display. • Reducing the data transmission frequency has the benefit of reducing the power consumption in the row and column drives. • Most of the pixel area required for the tandem and horizontal lines in previous patented solutions can be used for other purposes. • The area of the transparent electrode can be increased, and thus the transmittance of the display can be improved. • The display panel can be controlled using traditional display technology. • The display will be manufactured using twin crystal technology. • The display will be manufactured using MEMS technology. • The display will be manufactured using Field Emission Display (FED) technology. 116 200916986 • The hologram is converted to a one-dimensional conversion. • The hologram is converted to a two-dimensional transform. • Alternate circuit components (such as TFTs) may be fabricated in the space of the pixel matrix so that these circuit components can be used to replace some of the device startups when it is found to be faulty. The circuit components used. • * The wavefront emitted by the object will be reconstructed in one or more virtual observer windows (v〇w), and reconstruction of each single object point (OP) of one of the three-dimensional scenes (3D S) is only required A sub-holographic image (SH) is used as a subset of the entire hologram (IESLM) to be encoded on the SLM. • After a scene (3D S) is discretized into multiple object points (OP), for each visible object point of the 3D scene, the lens sub-image (the complex SH of SH will be encoded on the SLM) The complexity of the lens sub-image is determined by the following equation: ZL = exp { -i*[(π/λί ) * ( x2 + y2 )]}, where λ is the reference wavelength, f is the focal length, and x is y is the coordinate perpendicular to each other in the plane of the sub-image. • The sub-holographic image (SHP) of the 稜鏡 is determined in the holographic image plane (HE) to move the virtual observer window away from the optical axis. .  • The lens and cymbal are sub-images that are convoluted, which can also be represented by SH = SHl * SHP. . The space used to perform the holographic calculation may or may not be on the same substrate as the display substrate. . The circuit components that perform the decompression calculation are located between the pixels of the display. • The circuit component of the negative line reduction calculation is located outside the pixel matrix of the display but on the same plane. 117 200916986 • The cluster performs decompression calculations. • The cluster that performs the decompression calculation receives the data via the column and column lines of the display. • Each cluster that performs decompression calculations receives data via a parallel data bus. • Each cluster that performs decompression calculations receives data via a sequence data connection line. - • The method of using a full-image display. D. High-resolution display that can be decompressed on the same substrate A high-resolution display is used to display high-resolution image data that is first compressed using known data compression techniques and then transferred to the display's substrate. The circuit components, which then perform decompression on the received data, and then display the data on each pixel of the display. • The decompressed circuit components are located between the individual pixels of the display. • The decompressed circuit components are located outside of the display's pixel matrix but on the same substrate as the display. \ .  • The compressed data is transferred to the monitor to form the various clusters of the entire display. The cluster then decompresses the received data and then displays the data on each pixel of the regional cluster. • The general display data can be displayed. • The hologram display data can be displayed. • The space in which the compression calculation is performed may or may not be on the same substrate as the substrate of the display. 118 200916986 • The cluster that performs the decompression calculation receives the data via the column and column lines of the display. • Each cluster that performs decompression calculations receives data via a parallel data bus. • Each cluster that performs decompression calculations receives data via a sequence data connection line. • It is a very high resolution display. - • Each cluster performs decompression for 40 ms or less. . The hologram image calculation is performed after decompression. • At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located on the same substrate as the pixels of the spatial light modulator. • At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located on the same substrate as the pixels of the spatial light modulator, and these calculations themselves do not involve Fourier transform or Philippine The calculation of the Niel conversion. i'• At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located between the individual pixels of the spatial light modulator. • Calculations are performed in various areas scattered throughout the display to enable pixel encoding of the respective discrete areas for each discrete area. • A thin film transistor is included in the circuit assembly. • At least some of the circuit components have polysilicon in their active locations. 119 200916986 • At least some of the circuit components have active areas with continuous die 矽. • At least some of the circuit components have polysilicon in their active locations. • At least some of the circuit components have a single crystal 有效 in the effective location. • At least some of the circuit components have a valid area with a single die. - • At least some of the circuit components have an effective location with organic semiconductors. • Use a single crystal germanium substrate. . A glass substrate is used. • Video frame rate is at least approximately 25 Hz. • Only real-world image data will be transmitted to the monitor. • The image data contains light intensity and buffer corresponding data. • The holographic calculations performed are either immediate or near real-time operations. • The holographic calculations performed will be performed using the view method. • Sub-holographic images will be used for calculations. • The display will be manufactured using twin crystal technology. • The display will be manufactured using MEMS technology. 120 200916986 • The display will be manufactured using Field Emission Display (FED) technology. • The method of using a high-resolution display. E. At least a portion of the holographic image display that can be calculated on the same substrate by an additional processing unit for holographic conversion and encoding by a 3D rendering pipeline is used to determine a space. The encoding of the light modulator is calculated using a circuit component located on the same substrate as the pixel of the spatial light modulator, so the graphics subsystem's 3D rendering pipeline is integrated for holographic imaging. Additional processing units for conversion and encoding. • The holographic calculations performed will be performed using circuit components located between the various pixels of the display. • The holographic calculation is performed using circuit components located on the same substrate as the pixels of the display, other than the pixel matrix of the display. • At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located on the same substrate as the pixels of the spatial light modulator, and these calculations themselves do not involve Fourier transform or Philippine The calculation of the Niel conversion. • The calculation will be performed in various regions dispersed in the display to perform pixel coding of each respective dispersed region for each of the dispersed regions. • A thin film transistor is included in the circuit assembly. • Video frame rate is at least approximately 25 Hz. • Only real-world image data will be transmitted to the monitor. 121 200916986 • The image data contains light intensity and buffer corresponding data. • The holographic calculations performed are either immediate or near real-time operations. • The holographic calculations performed will be performed using the view method. • Use sub-holographic images for calculations. • The holographic operation is evenly distributed across the entire display surface. • The holographic calculation is divided into many congruent tiny parts (called clusters) that are tiled on the surface of the display. • According to the above concept, the holographic display of this case is a high-resolution display. • The diameter of a virtual observer window is approximately the diameter of the pupil of the eye or a larger diameter. • A monochrome image will be displayed. • The color image will be displayed. • When calculating the 値 of a pixel of a holographic image, only the 値 of a sub-segment of the original image is considered. • The light used to reconstruct the holographic image is not exactly the same across the display, but is exactly the same across the sub-sections of the display. . The hologram is converted to a one-dimensional conversion. . The hologram is converted to a two-dimensional transform. 122 200916986 • Alternate circuit components (such as TFTs) may be fabricated in the space of the pixel matrix so that these circuit components can be used to replace some of the device startups when it is found that some of the circuit components used for device startup are faulty. Circuit component 0 that will be used • The wavefront emitted by the object will be reconstructed in one or more virtual observer windows (VOW), and each single object point of a three-dimensional scene (3D S) The reconstruction of OP) requires only one sub-image as a subset of the entire hologram (ΗΣ51_μ) to be encoded on the SLM. • After a scene (3D S) is discretized into multiple object points (OPs), for each visible object point of the 3D scene, the complex sub-image of the lens (SHL) is encoded on the SLM, where The complexity of the lens sub-image is determined by the following equation: zL = exp { -i*[(π/λί ) * ( x2 + y2 )]}, where λ is the reference wavelength, f is the focal length, and x is y is the coordinate perpendicular to each other in the plane of the sub-holographic image. • The sub-holographic image (SHP) of the prism is determined in the holographic image plane (HE) to move the virtual observer window away from the optical axis. • The lens and 稜鏡 are sub-images that are convoluted, which can also be represented by SH = SHL * SHP. . Each sub-image (SH) is modulated with a uniformly distributed phase offset. Here, the phase offset # is different for each sub-image. • Multiple sub-images are superimposed to form the entire hologram. • The rendering of a holographic image produced by the computer used for reconstruction can be updated instantly or near instantaneously. * View tables are used in hologram calculations. • Individual object points can be generated anywhere in the frustoconic space used for reconstruction. 123 200916986 . The z-axis correspondence data for the first display wavelength is copied twice for use in the second and third display wavelengths. • A holographic image is calculated in parallel for each of the three display wavelengths. • Colors for two colors correspond to RGB content that is copied into individual memory segments to ensure that all three separations are independent. access. • The lens function and the 稜鏡 function for each display color perform a complex multiplication. • Each cluster of the display will have a random phase applied. • The calculated SLM code will use additional algorithms in the hologram display cluster for subsequent processing. • The method of using a hologram display. F. A holographic image display that can perform sequential omni-directional conversion of points in three-dimensional space and a calculation on the same substrate through a 3D pipeline (Pipeline) of the extended display card can be used to determine a spatial light modulation. The calculation of the encoding of the device is performed using circuit components located on the same substrate as the pixels of the spatial light modulator, such that sequential holographic conversion of points in the three-dimensional space is extended by a holographic computing pipeline. A 3D pipeline (Pipeline) to perform a hologram display. • The holographic calculations performed will be performed using circuit components located between the various pixels of the display. • The holographic calculation is performed using circuit components located outside the pixel matrix but on the same substrate as the display. • At least a portion of the metering shed used to determine the encoding of a spatial smoother is located on the same substrate as the pixel on which the spatial photoreceiver 124 200916986 is located, and the calculations themselves do not involve Fourier transforms. Or the calculation of the Fresnel transformation. • Calculations are performed in various areas scattered throughout the display to enable pixel encoding of the respective discrete areas for each discrete area. • A thin film transistor is included in the circuit assembly. • Video frame rate is at least approximately 25 Hz. • Only real-world image data will be transmitted to the monitor. • The image data contains light intensity and buffer corresponding data. • The holographic calculations performed are either immediate or near real-time operations. • The holographic calculations performed will be performed using the view method. • Use sub-holographic images for calculations. • The holographic operation is evenly distributed across the entire display surface. • The holographic calculation is divided into many congruent tiny parts (called clusters) that are tiled on the surface of the display. • According to the above concept, the holographic display of this case is a high-resolution display. • The diameter of a virtual observer window is approximately the diameter of the pupil of the eye or a larger diameter. • A monochrome image will be displayed. 125 200916986 • Color images will be displayed. • When calculating the 値 of a pixel of a holographic image, only the 値 of a sub-segment of the original image is considered. • The light used to reconstruct the holographic image is not exactly the same across the display, but is exactly the same across the sub-sections of the display. • The hologram is converted to a one-dimensional conversion. • The hologram is converted to a two-dimensional transform. • Alternate circuit components (such as TFTs) may be fabricated in the space of the pixel matrix so that these circuit components can be used to replace some of the device startups when it is found to be faulty. The circuit components used. • The wavefront emitted by the object will be reconstructed in one or more virtual observer windows (VOW) and the reconstruction of each single object point (OP) of one of the three-dimensional scenes (3D S) only needs one full time The image-like image (SH) is a subset of the entire hologram image (tESLM) to be encoded on the SLM. • After a scene (3D S) is discretized into multiple object points (0P), for each visible object point of the 3D scene, the complex sub-image of the lens (SHL) is encoded on the SLM, where The complex 値 of the lens sub-image will be determined by the following equation: A = exp { -i*[(πΜ ) * ( x2 + y2 )]} ' where λ is the reference wavelength, f is the focal length, and x and y It is a coordinate that is perpendicular to each other in the plane of the sub-holographic image. • The sub-holographic image (SHP) of the 稜鏡 is determined in the holographic image plane (HE) to move the virtual observer window away from the optical axis. • The lens and 稜鏡 are sub-images that are convoluted, which can also be represented by SH = SHL * SHP. 126 200916986 • Each sub-image (SH) is modulated with a uniformly distributed phase offset, where the phase shift is different for each sub-image. • Multiple sub-images are superimposed to form the entire hologram. • The rendering of a holographic image produced by the computer used for reconstruction can be updated instantly or near instantaneously. • View tables are used in hologram calculations. ♦ Individual object points can be generated anywhere in the frustoconic space used for reconstruction. • The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic transformation and encoding. • The Z-axis correspondence data for the first display wavelength is copied twice for the second and third display wavelengths. . A holographic image is computed in parallel for each of the three display wavelengths. • Colors for two colors correspond to RGB content that is copied into individual memory segments to ensure that all three separations are independently accessible. • The lens function and the 稜鏡 function for each display color perform a complex multiplication. • Each cluster of the display will have a random phase applied. • The calculated SLM code will use additional algorithms in the hologram display cluster for subsequent processing. . The hologram calculation can begin before there is a complete color correspondence and the Z-axis buffer data is available. 127 200916986 • The time required to perform a hologram calculation for each sub-image is less than the time of one image frame. • The time required to perform a hologram calculation for each sub-image is 17 ms or less. • Can be used for military purposes. • Each cluster of displays has its own view table for storing the encoding of the sub-holographic image it displays. • After reading the contents of the SH from the LUT, it calculates the difference between the currently displayed SHCSHU and the new SH (SHn). • Sequential holographic transformation of points in 3D space is performed by expanding the 3D pipeline of a PIC image display card, not limited to a specific type of SLM. • The method of using a hologram display. G. A computed holographic image display can be performed on the same substrate, a holographic display and can be randomly addressed at least in part to determine the calculation of the encoding of a spatial light modulator using the pixels located in the spatial light modulator The circuit components on the same substrate are executed so that the real-world image data used in the holographic calculation between successive real-world image frames is different, and the holographic display data will be sub-holographic image difference data and display memory. The form of the location data is transmitted to the holographic display of the holographic display cluster. • Sequential holographic conversion of points in a two-dimensional space is performed by expanding the 3D pipeline of a display card with a hologram pipeline. . The holographic calculations performed will be performed using circuit components located between the various pixels of the display. 128 200916986 • Full-image calculations are performed using circuit components located outside the pixel matrix but on the same substrate as the display. • At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located on the same substrate as the pixels of the spatial light modulator, and these calculations themselves do not involve Fourier transform or Philippine The calculation of the Niel conversion. • Calculations are performed in various areas scattered throughout the display to enable pixel encoding of the respective discrete areas for each discrete area. • A thin film transistor is included in the circuit assembly. • Video frame rate is at least approximately 25 Hz. . Only real-world image data will be transmitted to the display. • The image data contains light intensity and buffer corresponding data. • The holographic calculations performed are either immediate or near real-time operations. • The holographic calculations performed will be performed using the view method. • Each hologram image will be displayed. • The holographic operation is evenly distributed across the entire display surface. • The holographic calculation is divided into many congruent tiny parts (called clusters) that are tiled on the surface of the display. • According to the above concept, the holographic display of this case is a high-resolution display. 129 200916986 • The diameter of a virtual observer window is approximately the diameter of the pupil of the eye or a larger diameter. . A monochrome image will be displayed. • The color image will be displayed. . When calculating the 値 of one pixel of a holographic image, only the 値 of a sub-section of the original image is considered. • The light used to reconstruct the holographic image is not exactly the same across the display' but is exactly the same across the sub-sections of the display. • Full-image conversion stomach one-dimensional conversion. • The hologram is converted to a two-dimensional transform. • Alternate circuit components (such as TFTs) may be fabricated in the space of the pixel matrix so that these circuit components can be used to replace some of the device components that are used for device startup. The circuit components used. • The wavefront emitted by the object will be reconstructed in one or more virtual observer windows (vow), and the reconstruction of each single object point (OP) of one of the three-dimensional scenes (3D S) requires only one full The image (SH) is a subset of the entire hologram (fESLM) to be encoded on the SLM. • After a scene (3D S) is discretized into multiple object points (OPs), for each visible object point of the 3D scene, the complex sub-image of the lens (SHL) is encoded on the SLM. The complexity of the lens sub-image is determined by the following equation: zL = exp { -i*[(π/λί ) * ( x2 + y2 )]丨' where λ is the reference wavelength and f is the focal length, and y is the coordinate perpendicular to each other in the plane of the sub-holographic image. 130 200916986 • The sub-holographic image (SHP) of the 稜鏡 is determined in the holographic image plane (HE) to move the virtual observer window away from the optical axis. • The lens and 稜鏡 are sub-images that are convoluted, which can also be represented by SH = SHL * SHP. . Each sub-image (SH) is modulated with a uniformly distributed phase offset, where the phase shift is different for each sub-image. • Multiple sub-images are superimposed to form the entire hologram. . The rendering of the holographic image produced by the computer used for reconstruction can be updated instantly or near instantaneously. • Individual object points can be generated anywhere in the frustoconic space used for reconstruction. • The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic transformation and encoding. • The Z-axis correspondence data for the first display wavelength is copied twice for the second and third display wavelengths. • A holographic image is computed in parallel for each of the three display wavelengths. • Colors for two colors correspond to RGB content that is copied into individual memory segments to ensure that all three separations are independently accessible. • The lens function and the 稜鏡 function for each display color perform a complex multiplication. • Each cluster of the display will have a random phase applied. • The calculated SLM code will use additional algorithms in the hologram display cluster for subsequent processing. 131 200916986 • Can be used for military purposes. • Each hologram calculation unit receives image difference data. • If there is no difference between the displayed data of a particular cluster of consecutive sashes, or if there are only negligible differences that can be ignored, then there is no need to transfer the data to the cluster. • Each hologram conversion unit transmits 3D difference point image data relative to the reconstruction point or the point used to encode on the SLM. • There is a resolver in each hologram display cluster that decomposes the calculated holographic image display data into sub-holographic image data as well as size and position information, where the latter two 値 can be used to calculate the secondary &amp;; stomach § 彡 stomach # RAM address range, so that the sub-holographic image SH or SHD data can be written to the correct SLM pixel grid in the cluster. • Use a special random access memory (RAM) to write only the new SH or SHD on the input side, and the entire memory on the output side and write the complete information to the SLM. . • The method of using a hologram display. H. A display having an arithmetic function in the pixel space The arithmetic function is performed by a circuit component disposed on the same substrate as the pixel on which the display is located. • The arithmetic function is performed by circuit components disposed between the pixels of the display. . The arithmetic function will be performed by a circuit component disposed outside the pixel matrix but on the same substrate as the display. 0 132 200916986 • The display data on the display has a delay less than that of the circuit component disposed on the same substrate as the pixel of the display The arithmetic function performed is delayed if it is executed elsewhere. • These operations are graphical operations. • Can form part of a high-speed gaming device. • Can be used for military purposes. • The calculations are performed in various regions dispersed in the display to perform pixel encoding of the respective corresponding dispersed regions for each of the dispersed regions. • A thin film transistor is included in the circuit assembly. • At least some of the circuit components have polysilicon in their active locations. • At least some of the circuit components have active areas with continuous die. • At least some of the circuit components have polysilicon in their active locations. • At least some of the circuit components have a single crystal 有效 in the effective location. • Image data has a frame rate of at least approximately 25 Hz. • The operation (possibly parallel) is split into a number of congruent tiny parts that are tiled on the surface of the display. The display can be constructed by tiling many congruent clusters together. 133 200916986 • The display is a cartridge resolution display. • The display is an extremely high resolution display. . A color image will be displayed. • Color images in RGB format will be displayed. • The display will be manufactured using twin crystal technology. • There is an additional logic for the area to transfer existing calculated data, and this additional logic is also used to forward the original image to each cluster' so at least some common course and column lines can be eliminated. • How to use the monitor. I. The calculation of absorbing at least a portion of the code used to determine a spatial light modulator is performed using circuit components located on the same substrate as the pixels of the spatial light modulator, and ensures object points that are closer to the virtual observer window. A holographic display that obscures object points that are further away from the virtual viewer window along the same line of sight. • The calculation itself does not involve the calculation of Fourier transform or Philippine conversion. • The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then the holographic image encoding data is compressed by a known data compression technique and then transmitted to the circuit components on the display substrate. These circuit components then perform the function of decompressing the received data. • The graphics subsystem's 3D Rendering Pipeline integrates the 134 200916986 external processing unit for holographic conversion and encoding. • Sequential holographic transformation of points in 3D space is performed by augmenting the 3D pipeline (Pipeline) of the graphics card using a holographic calculation pipeline. • The real-world image data used in the hologram calculation is the difference between the contiguous real-space image frames, and the hologram display data is transmitted to the sub-image-like image difference data and the display memory location data. Each omnipresent display cluster. • "Absorption" is performed using calculations performed by circuit components disposed on the same substrate as the pixel matrix. • "Absorption" is implemented using calculations performed by circuit components disposed between pixels of the display. • The diameter of a virtual observer window is approximately the diameter of the pupil of the eye or a larger diameter. • VOW is divided into two or more sections. • The size of each vow segment is roughly the same size as the pupil size of the human eye. • Each vow segment is encoded with a different sub-holographic image. • “Absorption” is performed in the stage that constitutes the buffer corresponding data and light intensity corresponding data. • The method of using a hologram display. J. At least part of the function of the display card is used to determine the calculation of the encoding of a spatial light modulator. The computer is located on the same side as the surface of the spatial light modulator 135 200916986, and the function of the display card will be A holographic display that is executed by a circuit component on the same substrate on which the pixels of the facet are located. • The calculation itself does not involve the calculation of Fourier transform or Fresnel conversion. • The holographic image encoding data will be calculated outside the space occupied by the pixel matrix. Then the holographic image encoding data is used. The data compression technique compresses and then transmits to the circuit components on the display substrate, which in turn perform the function of decompressing the received data. . The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic conversion and encoding. • Sequential holographic transformation of points in 3D space is performed by augmenting the 3D pipeline (Pipeline) of the graphics card using a holographic calculation pipeline. • The real-world image data used in the hologram calculation is the difference between the contiguous real-space image frames, and the hologram display data is transmitted to the sub-image-like image difference data and the display memory location data. Each omnipresent display cluster. • The functionality of the display card is performed using circuit components disposed between the pixels of the display. • The function of the display card is performed using circuit components arranged outside the pixel matrix. • The functionality of the display card includes texture mapping. . The function of the display card includes rendering polygons. • The function of the display card includes translating the vertices into different coordinate systems. 136 200916986 • The functionality of the graphics card includes a programmable shader engine. • Display card features include oversampling and interpolation techniques to reduce overlays. • The function of the display card includes a very high precision color space. • The functionality of the display card includes 2D acceleration computing capabilities. • . The function of the display card includes the image buffer capability. • The features of the display card include the Animation Experts Group (MPEG) primitives. • The functionality of the display card includes the execution of operations involving matrix and vector operations. • The functionality of the display card includes the use of a 3D Rendering Pipeline performed by TFTs placed on the same substrate as the pixel matrix. • The method of using a hologram display. K.  2D (1) 3D conversion is used to determine at least a portion of the calculation of the encoding of a spatial light modulator using a circuit component located on the same substrate as the pixel of the spatial light modulator, and performing a 2D ^ 3D image conversion hologram display . • The calculation itself does not involve the calculation of Fourier transform or Fresnel transform. • The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then the holographic image encoding data is compressed by known data compression techniques and then transmitted to the circuit group on the display substrate 137 200916986 These circuit components then perform the function of decompressing the received data. • The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic transformation and encoding. . The sequential holographic transformation of each point in the three-dimensional space is performed by augmenting the 3D pipeline (Pipeline) of the display card using a holographic calculation pipeline. • The real-world image data used in the hologram calculation is the difference between the contiguous real-space image frames, and the hologram display data is transmitted to the sub-image-like image difference data and the display memory location data. Each omnipresent display cluster. . 2D (1) 3D image conversion is performed using circuit components disposed on the same substrate as the pixels of the display. • 2D (1) 3D image conversion is performed using circuit components that are not on the same substrate as the pixels on the display. • 2D (1) 3D image conversion is performed using circuit components disposed between pixels of the display. • 2D (1) 3D image conversion is performed using circuit components that are outside the pixel matrix but on the same substrate as the pixel. • 2D β 3D image conversion is performed using paired stereo images. • The display device calculates a two-dimensional (2D) image with corresponding buffer corresponding data based on the received data. • The circuit component used to perform the 2D (1) 3D conversion has access to a library of 138 200916986 containing a known set of 3D shapes. • The circuit component used to perform the 2D e 3D $ exchange has access to a library containing a known set of 2D shapes. In this database it may attempt to match the input 2D image data. • 2D 〇 3D image conversion is performed based on a single, non-automatic stereoscopic 2D image. • The method of using a hologram display. L. Video Conversation (3DSkypeTM) Full-image display will provide Internet Voice and holographic image protocol (VHIOIP) services. • At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located on the same substrate as the pixels of the spatial light modulator. • The calculation itself does not involve the calculation of Fourier transform or Fresnel transform. • The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then these holographic images are compressed using known data compression techniques and then transmitted to the circuit components on the display substrate. These circuit components then perform the function of decompressing the received data. • The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic transformation and encoding. • Sequential holographic transformation of points in 3D space is performed by augmenting the 3D pipeline (Pipeline) of the graphics card using a holographic calculation pipeline. • The real-world image data used for holographic calculation is the difference between 139 and 200916986 in each continuous real-world image frame, while the holographic display data will be sub-holographic image difference data and display memory and position data. The form is transmitted to each omnipresent display cluster. ι, mouth δ • VHIOIP point-to-point communication service can be provided. • File sharing can be provided. . Instant messaging service is available through a global network connected to it. • Communication services can be provided through a computer network connected to it. • File sharing services can be provided through a computer network connected to it. • Instant messaging service via a computer network connected to it. • Provides software for temporary, non-downloadable online use to allow users to use VHIOIP communication services. . Online software is available for download to allow users to use the VHIOIP communication service. . Provides access to the domain and domain database system to access holographic display data. • The method of using a hologram display. M. Code Compensation Compensation can be applied to holographic image data either during the encoding step or prior to the encoding step to provide a holographic display device that can see a clearer image. . At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located on the same substrate as the pixels of the spatial light modulator 140 200916986. • The calculation itself does not involve the calculation of Fourier transform or Philippine conversion. • The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then the holographic image encoding data is compressed by a known data compression technique and then transmitted to the circuit components on the display substrate. These circuit components then perform the function of decompressing the received data. • The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic transformation and encoding. • Sequential holographic transformation of points in 3D space is performed by augmenting the 3D pipeline (Pipeline) of the graphics card using a holographic calculation pipeline. • The real-world image data used in the hologram calculation is the difference between the contiguous real-space image frames, and the hologram display data is transmitted to the sub-image-like image difference data and the display memory location data. Each omnipresent display cluster. • "Compensation" is implemented using circuit components placed on the same substrate as the pixels of the display. . "Compensation" is implemented using circuit components disposed between the various pixels of the display. • “Compensation” is applied to the hologram image during the encoding step. • “Compensation” is applied to the hologram image before the encoding step. • Correction is applied to a scene that is dominated by bright tones and tends to be underexposed. • Correction is applied to a scene that is dominated by a dark tone and tends to be overexposed. 141 200916986 • The method of using the holographic display. N. Eye Tracking At least a portion of the calculations used to determine the encoding of a spatial light modulator are performed using circuit components located on the same substrate as the pixels of the spatial light modulator, and an eye-tracking holographic display is implemented. • The calculation itself does not involve the calculation of Fourier transform or Fresnel transform. • The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then the holographic image encoding data is compressed by a known data compression technique and then transmitted to the circuit components on the display substrate. These circuit components then perform the function of decompressing the received data. • The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic transformation and encoding. " . The sequential holographic transformation of each point in the three-dimensional space is performed by augmenting the 3D pipeline (Pipeline) using a holographic calculation pipeline. The following: The real-world image data used in the holographic calculation is the difference between each successive space image and the overall image of the sub-image difference and the holographic image of the image. Show clusters. Each • π] implements eye catching % for a single viewer. You can implement eye-catching _ for multiple viewers. Eye tracking will be used to limit the search shed, and then by detecting the eyes

142 200916986 範圍，接著再追蹤眼睛的位置來實施。 •會透過一個立體攝影機對用於執行眼睛位置辨識功能的計算模組提供一對立體影像。 .模組會回傳每個眼睛相對於一個固定點的X、y、及z軸座標。 .爲執行追蹤所需的運算會由位於與顯示器像素所在的同一基板上的電路組件來執行。 •爲執行追蹤所需的運算會由位於像素矩陣中的電路組件來執行。 • SLM面板上的全像編碼可以在面板的平面上移位。 •會透過將整個全像編碼內容在SLM上朝X或y軸方向移位來執行一個側向方向 的眼睛追蹤。 •追蹤可以透過使持續照亮SLM的光源隨著觀看者位置的改變而同步移動來執行。 •使用全像顯示器的方法。 〇·像差修正 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行，並且會實施像差修正的全像顯示器。 •計算的本身並不牽涉到傅立葉轉換或菲丨圼爾轉換的計算。 •全像影像編碼資料會在像素矩陣所佔用的空間以外的地方計算，然後再將這些全像 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 143 200916986 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉肖彳 外處理單元。 ‘‘、、額 •三維空間中的各個點的循序全像轉換會透過使用一個全像計算管線來擴充顯示卞自、 3D管線(Pipeline)來執行。 ' •用於進行全像計算中的真實空間影像資料是每一格連續真實空間影像框格之巧、 差，而全像顯示資料會以次全像影像差異資料以及顯示記憶位置資料的形式傳$ 全像顯示叢集。 ΰ •像差修正會利用與像素矩陣所在的同一基板上的電路組件來實施。 •像差修正會利用佈置在像素之間的電路組件來實施。 •像差可以透過空間光調制器的編碼來進行動態修正。 •所修正的像差是在一個凹凸透鏡陣列中的透鏡上的像差。 •所修正的像差是在一個2D透鏡陣列中的透鏡上的像差。 •會顯示各個次全像影像》 •會根據各個次全像影像產生一個總和全像影像。 •修正運算可以獨立於全像計算之外並行執行，一直到產生總和全像影像的步驟爲止。 •總和全像影像和像差修正對應可以一起進行調制。 144 200916986 •像差修正運算可以透過分析方式來執行。 •像差修正運算可以利用檢視表（LUT)來執行。 •使用全像顯示器的方法。 ρ·光斑修正 至少有一部分用來決定一個空間光調制器的編碼的計算會利用位於和空間光調制器的 像素所在的同一基板上的電路組件來執行，並且會執行光斑修正的全像顯示器。 .計算的本身並不牽涉到傅立葉轉換或菲涅爾轉換的計算。 •全像影像編碼資料會在像素矩陣所佔用的空間以外的地方計算，然後再將這些全像 影像編碼資料利用已知的資料壓縮技術進行壓縮，之後再傳輸到顯示器基板上的電路組 件上，這些電路組件接著再執行將所接收到的資料解壓縮的功能。 •圖形子系統的3D著色管線(Rendering Pipeline)整合有用於進行全像轉換和編碼的額 外處理單元。 •三維空間中的各個點的循序全像轉換會透過使用一個全像計算管線來擴充顯示卡的 3D管線(Pipeline)來執行。 •用於進行全像計算中的真實空間影像資料是每一格連續真實空間影像框格之間的 差’而全像顯示資料會以次全像影像差異資料以及顯示記憶位置資料的形式傳送給各個 全像顯示叢集。 •光斑修正會利用與像素矩陣所在的同一基板上的電路組件來實施。 145 200916986 •光斑修正會利用佈置在像素之間的電路組件來實施。 •光斑可以透過空間光調制器的編碼來進行動態修正。 •會顯示各個次全像影像$ •會根據各個次全像影像產生一個總和全像影像。 .光斑修正運算可以獨立於全像it算之外並行執行，一直到產生總和全像影像的步驟 爲止。 •總和全像影像和光斑修正對應可以一起進行調制。 •光斑修正運算可以透過分析方式來執行。 •光斑修正運算可以利用檢視表（LUT)來執行。 •使用全像顯示器的方法。 Q.全像顯示器的數位版權管理技術（DRM)解碼 解碼和全像影像計算會利用位於像素矩陣所在的基板上的電路組件來執行的全像顯示 裝置。 •解碼和全像影像計算會利用分散在整個像素矩陣基板上的電路組件以—個分散的槪 念來執行。 •解碼和全像影像計算會利用位於像素矩陣中的電路組件來執行。 146 200916986 •解碼和全像影像計算會利用位於像素矩陣以外、但在與像素矩陣所在的同一基板上 的電路組件來執行。 •麵上^^會有任何—娜一的位置可以娜肺經過麵馬的貪料。 •整個面板上的不同區位都使用有不同的解碼密鑰。 •使用全像顯示器的方法。 R. 2D顯示器的數位版權管理技術（DRM)解碼 解碼計算會利用分散在整個像素矩陣基板上的電路組件以一個分散的槪念來執行的2D 顯示裝置。 .解碼計算會利用位於像素矩陣中的電路組件以一個分散的槪念來執行。 •解碼計算會利用位於像素矩陣以外、但在與像素矩陣所在的同一基板上的電路組件 以一個分散的槪念來執行。 •基板上不會有任何一個單一的位置可以截取所有經過解碼的資料。 •整個基板上的不同區位都使用有不同的解碼密鑰。 •使用顯示器的方法。 解碼計算會利用位於顯示器基板的一個單一區位中的電路組件來執行的2D顯示裝 置。 147 200916986 •這些電路組件可以在像素矩陣內部。 •這些電路組件可以在像素矩陣以外。 •使用顯示器的方法。 S. 在與顯示器以實體線路連接的硬體中執行軟體應用程式 一個可以利用軟體來執行的應用程式改爲利用分散在一個SLM面板的整個基板上的 電路組件在硬體中執行的顯示裝置。 •這個顯示器是一個2D顯示器。 •這個顯示器是一個全像顯示器。 •應用程式會利用位於顯示器的各個像素之間的電路組件來執行。 •應用程式會利用位於顯示器的像素矩陣以外的電路組件來執行。 •使用顯示器的方法。 T. 採用多個微稜鏡的可變光束轉向 一位觀看者或多位觀看者可以利用一個可以控制光朿轉向的微稜鏡陣列來進行追蹤的 全像顯示器。 •可以利用縱列的兩個微稜鏡陣列來進行兩個維度的光線轉向。 •這些稜鏡是液態微稜鏡。 148 200916986 .可以減少透鏡像差的光學效應。 • vow設在一位觀看者或多位觀看者眼睛的位置。 •加設在稜鏡陣列之前或之後的一個聚焦的手段將可以協助將光線匯集到vow上。 •稜鏡並非都有相同的偏導角度。 •稜鏡並非都有相同的偏導角度使離開稜鏡陣列的光線可以大致匯集到vow處。 •稜鏡角度的計算會在位於SLM的基板上的運算電路組件中執行。 .稜鏡角度的計算會在佈置於稜鏡陣列的基板上的運算電路, 組件中執行。 • SLM的基板也可以用來做爲稜鏡陣列的基板。 •會施加一個“相位修正”針對因稜鏡陣列所造成的相位相位不連續的現象提供補償。 .相位修正可以由SLM執行。 •全像影像是在一個投影式設備中產生，這裡的投影牽涉到在稜鏡陣列上構成一個 SLM影像，而所需的3D景象的重建則發生在VOW的前方。 •針對稜鏡陣列的相位補償會在在稜鏡陣列上構成SLM影像時提供。 •針對稜鏡陣列的相位補償會由設置在靠近稜鏡陣列的一個額外的SLM提供。 • SLM可能是可以透光的而稜鏡陣列則可以反射。 • SLM可能是可以反射的而稜鏡陣列則可以透光。 •使用全像顯示器的方法。 149 200916986 【圖式簡單說明】 圖1疋用來顯7F全像影像的㈣傳輸速率遠高於原始真實空間資料的資料 傳輸速率的說明圖； 圖2疋^中的SLM的—部份與—個可以在像素矩陣的空間中執行 王像的SLM的-部份在構造和性能雜方面的比較圖； 圖3是-個可以在像素矩陣的空間中執行全像計算的slm的一部份的構 造圖， 圖4是-個可以在像素矩_空間中執行解壓縮計算關於全像資料顯示 的SLM的一部份的構造圖； 圖5是-個可以在像素矩陣的空間中執行解壓縮計算以用於傳統加顯示 資料顯示的SLM的-部份的構造圖； 圖6 &gt;〇’、貞不TFT的製造過程中的各個情況的說明圖； 圖7是顯示TFT的製造擁巾的各鑛⑽說明圖； 圖8是顯錄縣翻的—麵辟計來«全像影像的方法的說明圖； 圖9是齡_本發明的—個顯示設計來重建全像影像的方法的說明圖； 圖10疋以先_專利為根據的一個傳統主動矩陣式液晶顯示農置的一般 構造的透視圖； 圖11包括顯不以本發明的—個顯示設計為根據的全像顯示器的一個主動 矩陣式基板的各個製造步驟的說明圖； 圖12包括顯示圖u 圖； 圖13包括顯示圖12 中的主動矩陣式基板的各個進—步製造步驟的說明 中的主動矩陣式基板的各健_步製造步驟的說明 150 200916986 圖； 說明圖； 圖14是-録各個分散且任意的位置上表現各個物件關全像顯示器的 圖15是可能提供在以本發個齡設計為根據的全像顯示器中 形計算中的功能單元的說明圖； 、回 圖16是用於以本發_—個顯示設計為根據的全像顯示器中的次 像SH的檢視表的說明圖； 〜 圖17是用於以本發明的一個顯示設計為根據的全像顯示器中進行 換與編碼的額外處理單元的說圖； 王 圖18是顯雜以本發個顯示設計為根據的全像顯㈣中，如果採用 次全像影像爾貞雜她、⑽输陳㈣）的說明圖； 圖19、是顯示-個在時^時顯示的景象、另—個在時間糾時顯示的景 象、以及兩者之間的景象差別的說明圖； 圖20是顯示具備有可編址㈣傳輸能力的以本發明的—個顯示設計為根 據的全像顯示裝置的說明圖； 圖21顯示-個會在其中計算以本發 _ 個顯不没計為根據的全像顯示 器中的電晶體數量的試算表的一部份； 圖22是圖21中所示的試算表的其餘部份； ^是以繼输祕集設計 資料 圖24是鳴她的—侧爾㈣軸繼為根據的顯示 151 200916986 所採取的路徑的說明圖； 、 ;項顯示的運算上的各項計算可能在像素矩陣的空間中執行 的了以顯不傳統2D顯示資料、或全像顯示資料的—個SLM的一部份的 構造圖； =乂先_專利為根據、用來產生次全像影像的方法的說明圖； =如根據本發明的—個顯示設計來重建全像影像的方法的說明圖； 疋以本發明的-個顯示設計為根據的面祕貼的構造圖； 疋關於則欠的幾何上的考量的的說明圖； 圖30是關於“吸收，，的 的成何上的考量的的說明圖； 圖31是根據本發明的—個 ，，&quot;員不δ又叶來處理吸收現象的方法的說明圖； 圖32是根據本發明的—個 頜不§又计來處理吸收現象的方法的說明圖； 圃心是以用於本發明一 一 —個顯示設計的全像顯示裝置為根據的顯示資料 所採取的路徑的說明圖； 圖財是根據本發明的—個 者視窗來追縱-個或多個你又计，利用可控制的稜鏡透過移動虛擬觀察 之用者的方法的說明圖； 元件符號簡單說明： 601破璃基板 6〇2氧化矽 603非結晶矽薄膜 604耐蝕遮罩 605薄膜 152 200916986 606含Ni層 607多晶SiuxGex區域 608多晶石夕區域 609作用層 610作用層 711,712,713閘極絕緣薄膜 714,715,716 閘極 717,718,719緊致陽極氧化物薄膜 720遮罩 721汲極區域 722源極區域 723 LDD區域 724通道區域 725源極區域 726汲極區域 727 LDD區域 728通道區域 729耐蝕遮罩 730源極區域 731汲極區域 732LDD區域 733通道區域 734中介層絕緣薄膜 153 200916986 735,736,737 源極 738,739 汲極 740中介層絕緣薄膜 741黑色遮罩 742絕緣層薄膜 743像素電極 VOW虛擬觀察者視窗 SLM空間光調制器 OP 物件點 HAE全像顯示裝置 SH 次全像影像 101主基板 102相對基板 103間隔空間 104像素電極 105開/關裝置 106顯示部份 107週邊驅動部份 1101基板 1102薄膜 1103遮罩絕緣薄膜 200916986 1104促進劑薄膜 1105區域 1106區域 1107水平生長區域 1108水平生長區域 1109,1110,1111島狀的半導體層 1112閘極絕緣薄膜 1113,1114,1115 閘極 1116,1117,1118陽極化薄膜 1119,1120,1121電極的正下方 1122耐蝕遮罩 1123,1124,1125,1126 η 型區域 1127耐蝕遮罩 1128,1129 ρ 型區域 1130,1131,1132 側壁 113 3遮罩 1134,1138源極區域 1135,1139汲極區域 1136,1140低濃度雜質區域 1137原生的通道成形區域 1141通道成形區域 1142埘蝕遮罩 155 200916986 1147 Ti薄膜 1148,1149,1150矽化物區域 1151,1152,1153島狀的型樣 1154中介層絕緣薄膜 1155,1156,1157 源極配線 1158,1159汲極配線 1160絕緣層 1161黑色遮罩 1162絕緣中介層薄膜 1163像素電極 1501,1502 Z軸對應（映圖）資料的拷貝 1503,1504,1505進行複乘法 1506,1507,1508套用隨機的相位 1509,1510,1511調制出個別的全像影像 CORDIC位數對照方法，Voider運算法,CORDIC也就是“座標 轉動數位電腦”的縮寫 156142 200916986 range, then track the position of the eye to implement. • A pair of stereo images will be provided to the computing module used to perform the eye position recognition function through a stereo camera. The module will return the X, y, and z axis coordinates of each eye relative to a fixed point. The operations required to perform the tracking are performed by circuit components located on the same substrate as the display pixels. • The operations required to perform the trace are performed by the circuit components located in the pixel matrix. • The hologram code on the SLM panel can be shifted on the plane of the panel. • Perform a lateral direction eye tracking by shifting the entire hologram encoded content on the SLM toward the X or y axis. • Tracking can be performed by causing the light source that continuously illuminates the SLM to move synchronously as the position of the viewer changes. • The method of using a hologram display. 〇·Aberration correction At least a portion of the calculation used to determine the encoding of a spatial light modulator is performed using circuit components located on the same substrate as the pixels of the spatial light modulator, and a hologram of aberration correction is implemented. monitor. • The calculation itself does not involve the calculation of Fourier transform or Fischer transformation. • The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then the holographic image encoding data is compressed by a known data compression technique and then transmitted to the circuit components on the display substrate. These circuit components then perform the function of decompressing the received data. 143 200916986 • The 3D Shading Pipeline of the graphics subsystem is integrated with a processing unit for holographic processing. ‘‘, 额 • The sequential holographic transformation of each point in the three-dimensional space is performed by using a holographic calculation pipeline to augment the display, 3D pipeline (Pipeline). 'The real-world image data used for holographic calculation is the coincidence and difference of each continuous real-world image sash, and the holographic display data will be transmitted in the form of sub-holographic image difference data and display memory location data. $ Full image shows clusters. • Aberration correction is performed using circuit components on the same substrate as the pixel matrix. • Aberration correction is implemented using circuit components arranged between pixels. • Aberrations can be dynamically corrected by the encoding of the spatial light modulator. • The aberration corrected is the aberration on the lens in a lenticular lens array. • The aberration corrected is the aberration on the lens in a 2D lens array. • Each hologram image will be displayed • A holographic image will be generated based on each hologram image. • Correction operations can be performed in parallel independently of holographic calculations, up to the step of generating a holographic image. • The total hologram and aberration correction can be modulated together. 144 200916986 • The aberration correction operation can be performed by analysis. • The aberration correction operation can be performed using a view table (LUT). • The method of using a hologram display. ρ. Spot Correction At least a portion of the calculation used to determine the encoding of a spatial light modulator is performed using circuit components located on the same substrate as the pixels of the spatial light modulator, and a holographic display that performs spot correction is performed. The calculation itself does not involve the calculation of Fourier transform or Fresnel transform. • The holographic image encoding data is calculated outside the space occupied by the pixel matrix, and then the holographic image encoding data is compressed by a known data compression technique and then transmitted to the circuit components on the display substrate. These circuit components then perform the function of decompressing the received data. • The graphics subsystem's 3D Rendering Pipeline incorporates additional processing units for holographic transformation and encoding. • Sequential holographic transformation of points in 3D space is performed by augmenting the 3D pipeline (Pipeline) of the graphics card using a holographic calculation pipeline. • The real-world image data used in the holographic calculation is the difference between the contiguous real-world image frames of each cell, and the holographic display data is transmitted to the sub-image-like image difference data and the display memory location data. Each omnipresent display cluster. • Spot correction is performed using circuit components on the same substrate as the pixel matrix. 145 200916986 • Spot correction is implemented using circuit components placed between pixels. • The spot can be dynamically corrected by the encoding of the spatial light modulator. • Each hologram image will be displayed. • A holographic image will be generated based on each hologram image. The spot correction operation can be performed in parallel independently of the hologram it, until the step of generating a holographic image is generated. • The sum hologram and spot correction can be modulated together. • The spot correction operation can be performed by analysis. • The spot correction operation can be performed using a view table (LUT). • The method of using a hologram display. Q. Digital Rights Management Technology (DRM) decoding of holographic displays Decoding and holographic image computing utilizes a circuit component located on a substrate on which the pixel matrix is located to perform a holographic display device. • Decoding and holographic image calculations are performed with a discrete pattern of circuit components scattered throughout the pixel matrix substrate. • Decoding and holographic image calculations are performed using circuit components located in the pixel matrix. 146 200916986 • Decoding and holographic image calculations are performed using circuit components located outside of the pixel matrix but on the same substrate as the pixel matrix. • There will be any ^^ on the face - the position of Na is able to pass through the face of the horse. • Different decoding keys are used for different locations on the entire panel. • The method of using a hologram display. Digital Rights Management Technology (DRM) Decoding of R. 2D Display The decoding calculation utilizes a circuit component dispersed throughout the pixel matrix substrate to perform a 2D display device with a discrete commemoration. The decoding calculation is performed with a discrete commemoration using the circuit components located in the pixel matrix. • The decoding calculation is performed with a discrete commemoration of the circuit components located outside the pixel matrix but on the same substrate as the pixel matrix. • There will be no single location on the substrate to capture all decoded data. • Different decoding keys are used for different locations on the entire substrate. • How to use the monitor. The decoding calculation utilizes a 2D display device that is executed by circuit components located in a single location of the display substrate. 147 200916986 • These circuit components can be inside the pixel matrix. • These circuit components can be outside the pixel matrix. • How to use the monitor. S. Executing a software application in a hardware connected to a display with a physical line An application that can be executed by software is instead a display device that is executed in hardware by using a circuit component dispersed on the entire substrate of one SLM panel. • This monitor is a 2D monitor. • This monitor is a hologram display. • The application will execute using circuit components located between the various pixels of the display. • The application will execute using circuit components located outside of the display's pixel matrix. • How to use the monitor. T. Variable beam steering with multiple micro-turns A one-viewer or multiple viewers can use a micro-image array that controls the pupil's steering for tracking. • Two-dimensional arrays of light can be used to make two dimensions of light steering. • These defects are liquid micro-caries. 148 200916986 . Can reduce the optical effect of lens aberration. • vow is located in the eye of one viewer or multiple viewers. • A focusing means added before or after the 稜鏡 array will assist in the collection of light onto the vow. • 稜鏡 does not all have the same deflection angle. • The 稜鏡 does not all have the same deflectance angle so that the light exiting the 稜鏡 array can be roughly summed to the vow. • The calculation of the 稜鏡 angle is performed in an arithmetic circuit component located on the substrate of the SLM. The calculation of the 稜鏡 angle is performed in an arithmetic circuit, component disposed on the substrate of the 稜鏡 array. • The SLM substrate can also be used as a substrate for the 稜鏡 array. • A “phase correction” is applied to compensate for the phase phase discontinuity caused by the 稜鏡 array. The phase correction can be performed by the SLM. • The holographic image is produced in a projection device where the projection involves constructing an SLM image on the 稜鏡 array, and the reconstruction of the required 3D scene occurs in front of the VOW. • Phase compensation for the 稜鏡 array is provided when the SLM image is formed on the 稜鏡 array. • Phase compensation for the 稜鏡 array is provided by an additional SLM located close to the 稜鏡 array. • The SLM may be light transmissive and the 稜鏡 array can reflect. • The SLM may be reflective and the 稜鏡 array may be transparent. • The method of using a hologram display. 149 200916986 [Simple diagram of the diagram] Figure 1 is used to display 7F hologram image (4) transmission rate is much higher than the original real-time data data transmission rate; Figure 2 S ^ SLM - part and A comparison of the structure and performance of the SLM that can perform the image of the image in the space of the pixel matrix; Figure 3 is a part of the slm that can perform the holographic calculation in the space of the pixel matrix Construction diagram, FIG. 4 is a structural diagram of a portion of the SLM that can perform decompression calculation in the pixel moment_space for omnidirectional data display; FIG. 5 is a graph that can perform decompression calculation in the space of the pixel matrix FIG. 6 is an explanatory view of each case in the manufacturing process of the SLM for the display of the conventional display data; FIG. 6 is an explanatory view showing each case in the manufacturing process of the TFT; Mine (10) explanatory diagram; FIG. 8 is an explanatory diagram of a method for displaying a holographic image by a recorded county; FIG. 9 is an explanatory diagram of a method for displaying a holographic image by displaying the design of the invention. Figure 10 is a traditional active moment based on the patent A perspective view of a general configuration of an array liquid crystal display farm; FIG. 11 includes an illustration of various manufacturing steps of an active matrix substrate showing a holographic display based on a display design of the present invention; Figure 14 is a diagram showing the steps of the various steps of the active matrix substrate in the description of the various step-by-step manufacturing steps of the active matrix substrate of Figure 12; - Figure 15 is a diagram showing the functional units in the holographic display of the holographic display based on the design of the present invention; FIG. Is an explanatory view of a view table for the secondary image SH in the hologram display based on the present invention; FIG. 17 is for use in a hologram display based on a display design of the present invention. Change the picture with the additional processing unit of the code; Wang Figure 18 is a holographic display based on the display design of this release (4), if the sub-holographic image is used, noisy her, (10) loses Chen (4) FIG. 19 is an explanatory diagram showing a scene displayed at time ^, a scene displayed at time correction, and a scene difference between the two; FIG. 20 is a display capable of editing An illustration of a holographic display device based on the display design of the present invention based on the transmission capability of the present invention; FIG. 21 shows a hologram display in which the calculation is based on the sensation of the present invention. A part of the spreadsheet of the number of transistors; Figure 22 is the rest of the spreadsheet shown in Figure 21; ^ is based on the design data of the secret collection. Figure 24 is the basis of her-side (four) axis. The display of the path taken by 151 200916986; the calculation of each item on the operation of the item display may be performed in the space of the pixel matrix to display the data of the traditional 2D display data or the hologram display. a partial structural diagram; = 乂 patent _ patent is an explanatory diagram of a method for generating a sub-holographic image; = an explanatory diagram of a method for reconstructing a holographic image according to a display design according to the present invention;疋Based on the display design of the present invention FIG. 30 is an explanatory diagram of the consideration of "absorption," and FIG. 30 is an explanatory diagram of the consideration of "absorption," and FIG. FIG. 32 is an explanatory diagram of a method for processing an absorption phenomenon according to the present invention; FIG. 32 is an explanatory diagram of a method for processing an absorption phenomenon according to the present invention; The invention provides a description image of the path taken by the display image based on the holographic display device of the display design; the image is based on the window of the present invention to track one or more Description of the method of controlling the user through the virtual observation; Brief description of the component symbol: 601 glass substrate 6〇2 矽 矽 603 amorphous 矽 film 604 corrosion resistant mask 605 film 152 200916986 606 Ni layer 607 Polycrystalline SiuxGex region 608 polylithic region 609 active layer 610 active layer 711, 712, 713 gate insulating film 714, 715, 716 gate 717, 718, 719 compact anodic oxide film 720 mask 721 drain region 722 source region 723 LDD region 724 Channel region 725 source region 726 drain region 727 LDD region 728 channel region 729 corrosion resistant mask 730 source region 731 drain region 732LDD region 733 channel region 734 interposer insulating film 153 200916986 735,736,737 source 738,739 bungee 740 interposer insulation Film 741 Black Mask 742 Insulation Film 743 Pixel Electrode VOW Virtual Observer Window SLM Spatial Light Modulator OP Object Point HAE Full Image Display Device SH Secondary Image 101 Main Substrate 102 Relative Substrate 103 Space 104 Radiation Closing device 106 display portion 107 peripheral driving portion 1101 substrate 1102 film 1103 mask insulating film 200916986 1104 accelerator film 1105 region 1106 region 1107 horizontal growth region 1108 horizontal growth region 1109, 1110, 1111 island-like semiconductor layer 1112 gate Insulating film 1113, 1114, 1115 gate 1116, 1117, 1118 anodized film 1119, 1120, 1121 directly below the electrode 1122 corrosion resistant mask 1123, 1124, 1125, 1126 n-type region 1127 corrosion resistant mask 1128, 1129 ρ-type region 1130, 1131, 1132 sidewall 113 3 mask 1134, 1138 source region 1135, 1139 drain region 1136, 1140 low concentration impurity region Domain 1137 native channel forming region 1141 channel forming region 1142 etched mask 155 200916986 1147 Ti film 1148, 1149, 1150 bismuth region 1151, 1152, 1153 island-like pattern 1154 interposer insulating film 1155, 1156, 1157 source Pole wiring 1158, 1159 bungee wiring 1160 insulating layer 1161 black mask 1162 insulating interposer film 1163 pixel electrode 1501, 1502 Z axis corresponding (map) copy of the data 1503, 1504, 1505 for multiplication 1506, 1507, 1508 application Random phase 1509, 1510, 1511 modulate the individual holographic image CORDIC digit comparison method, Voider algorithm, CORDIC is also the abbreviation of "coordinate rotary digital computer" 156

Claims (1)

200916986 、申請專利範圍： 1. 一種可齡、_解_的高解析度影«高崎度顯㈣,料 包括無數像素，這些像素⑽—個额上，射的電馳件也位於與各個 象素斤在的同錄板± ’其巾的高解影像資料採用已知 的資料壓縮技術進行壓縮，二、+，而 . 縮亚由_電路組件接收；電路組件的作用可執 行解壓縮§f算，並提供經過解壓 ，、敗缩的向解析度影像資料，由顯示器上的各 個像素進行後續的顯示。 ==Γ項所述之高解析度顯示器，其”腦運算功能由佈 在”顯不③的各個像素所在的同—基板上的電路組件執行。 3.如申請專利範圍第Μ 2項所述之高解析度顯 fr會利用分揭卿崎_—-«1= 軟體來執行的應歸式已經改為_分散在 項所述之高解析度顯示器，射—個可以利用 4.如申請專利範圍前述任— 組件在硬體中執行 整個顯示B像素基板上的電路 5.如申請專利範圍前述任— ^ 紐瓦 項所奴肖解析度齡ϋ，財可以用來執杆 解壓縮計算的電馳件佈置在顯示騎各轉素之間。 157 200916986 2申請她购^雜—顧㈣嫩_，其心 夕h 打的電路組件佈置在包含顯示器的各個像素的空^ 外，但在無轉的像細在㈣—基板上。 門之 ^如申糊範⑼到5項任_娜之高解析翻示器其 =的㈣會傳朗顯示器的各觸集上，每個叢集都是整個顯示器不可 刀割的-部份，母個叢集接著會對所接收到的轉執行—個解壓縮功能， 然後再由每個叢集的各個像素顯示經過解壓縮的資料。 8.如申請梅瞻任-恤彻，其巾賴示器可以 顯示一般顯示資料。 9. 如申請專繼圍前述任-項所述之高解析度顯示器，其中的顯示訊以 顯示全像顯示資料。 1〇·如申請專利範圍前述任-項所述之高解析度顯示器，其中用來執行壓縮 計算的空間係與顯示器的各個像素位於同—基板上。 11.如申請專利範圍第1到 ^頁所述之南解析度顯示器，其中用來 執行壓縮計算的空間與顯示器的各個像素在不同—義板上 158 200916986 12.如申明專利範圍第7到u項任一項所述之高解析度顯示器，其中用於 進行解義計算哺齡軸顯4的铜及縱舰路接收資料。 13. 如申„月專利範圍第7和項任一項所述之高解析度顯*器，其中每個 用於解壓&amp; 3博的叢集會經由—個並職料隨排接收資料。 14. 士申叫專利細第7 ^JU項任一項所述之高解析度顯示器，其中每個 用於解壓縮計算的叢齡軸—辦贿料連線接收資料。 瓜如申請專繼圍前述任—項職之高解·顯示器，是—飾高解析度 的顯示器。 16.如申請專利範圍第7 壓縮會由每個叢集在40 到15項任一項所述之高解析度顯示器，其中的解 ms或以下的時間内執行。 17.如申請專利範圍第9 全像影像計算會在解壓縮 到16項任一項所述之高解析度顯示器 之後執行。 ’其中的 18.如申請專利範圍第1?項所述之高解析度顯 間光調制H的編碼所執行的計算 示器，其中用來決定 '個空 會利用佈置在與空間光調制器的各個像素 159 200916986 所在的同一基板上的電路組件來執行。 19.如申請專利範圍第18項所述之高解析度顯示器，其中用來決定—個空 間光調制器的編碼所執行的計算並不牵涉到傅立葉轉換（⑽加 ffooO或菲埋爾轉換（如咖bans—)的計算。 &gt; 20.如申請專利範圍第18或19項所述之高解析度顯示器，其中用來争定— 個空間光議㈣彳顺算細佈置她賊㈣的各I 素之間的電路組件來執行。 2央L如申請專利範圍第18至，1 2〇項任一項所述之高解析度顯示器，其中用 ^=:個空間光調㈣的編碼所執行的計算會在分散在顯示器中的 域執行，讀針議繼域，進行蝴咖咖 22.如申請專利範圍前述任一項所述 括有薄臈電晶體。 之高解析度顯示器，其中的 電路組件包 其中至少有一部份 23.如申請專利範圍前述任一項所述之高解析度顯示器 的電路組件的有效區位由多晶矽所構成。 160 200916986 24.如申請專利綱前雜-項所述之高解析度顯示器 ，其中至少有一部份 的電路組件的有效區位由連續晶粒石夕所構成。 25.如申請專利範圍前述任-項所述之高解析度顯示器，其中至少有一部份 的電路組件的有效區位由多晶石夕鍺所構成。 26.如申請專利範圍前述任—項所述之高解析度顯示器，其中至少有一部 份的電路組件的有效區位由單晶矽所構成。 27. 如中轉她_述任_項所述之高解析度顯示器，其中至少有一部 份的電路組件的有效區位由單體晶粒矽所構成。 28. 如申請專利範圍前述任—項所述之高解析度顯示器，其中至少有一部 份的電路組件的有效區位由有機半導體所構成。 29. 如申請專利範圍前述任一項所述之高解析度顯示器，其中的基板採用 單晶石夕。 30.如申請專利範圍第19項所述之高解析度顯示器，其中的顯示器在製造 上採用矽晶液晶技術。 161 200916986 31. 如申請專利範圍第1到28項任一項所述之高解析度顯示器，其中的基 板採用玻璃。 32. 如申請專利範圍前述任一項所述之高解析度顯示器，其中的視訊影格 速率至少約為25 Hz。 33. 如申請專利範圍第9到32項任一項所述之高解析度顯示器，其中只有 真實空間影像資料會被傳輸給顯示器。 34. 如申請專利範圍第33項所述之高解析度顯示器，其中的真實空間影像 資料由光強度及縱深對應資料所構成。 35. 如申請專利範圍第17到34項任一項所述之高解析度顯示器，其中的 全像計算為即時或近即時運算。 36. 如申請專利範圍第17到35項任一項所述之高解析度顯示器，其中所 進行的全像計算會採用檢視表方法來執行。 37. 如申請專利範圍第17到36項任一項所述之高解析度顯示器，其中採 用次全像影像來進行全像計算。 162 200916986 38. 如申請專利範圍第1到29項任一項所述之高解析度顯示器，其中的顯 示器在製造上採用MEMS技術。 39. 如申請專利範圍第1到29項任一項所述之高解析度顯示器，其中顯 示器在製造上將採用場發射顯示器（FED)技術。 40. —種使用前述任一專利主張所述的高解析度顯示器來顯示經過解壓縮 的高解析度影像資料的方法，包括下列步驟： (a) 壓縮高解析度影像資料； (b) 傳輸經過壓縮的高解析度影像資料； (c) 接收經過壓縮的高解析度影像資料； (d) 對經過壓縮的高解析度影像資料進行解壓縮；以及 (e) 將經過解壓縮的高解析度影像資料顯示在顯示器上。 163200916986, the scope of application for patents: 1. A high-resolution shadow of the age, _ solution _ [高崎度显(4), including a myriad of pixels, these pixels (10) - the amount of the upper part, the shooting of the electrical components are also located in each pixel The same recording board of the kg ± 'the high resolution image data of the towel is compressed by the known data compression technology, two, +, and the shrinking sub-circuit is received by the circuit component; the function of the circuit component can be decompressed §f And provide decompressed, retracted to the resolution image data, followed by the display of each pixel on the display. The high-resolution display of the == item, whose "brain computing function" is performed by a circuit component disposed on the same substrate as the respective pixels of the display. 3. If the high-resolution FR described in item 2 of the patent application scope is used, the categorization performed by the squad __-«1= software has been changed to _ dispersing the high resolution described in the item. The display can be used as follows. 4. As described in the patent application scope, the component performs the entire circuit on the B-pixel substrate in the hardware. 5. As described in the patent application, the nucleus is not included in the scope of the invention. The money can be used to implement the decompression calculation of the electrical components arranged between the display riders. 157 200916986 2 Applying for her purchase of the miscellaneous - Gu (four) tender _, the circuit components of the hearth are arranged outside the space of each pixel containing the display, but the image without rotation is fine on the (four)-substrate. The door of the door ^ such as the application of the paste (9) to the five items of the _ Na high resolution flip-flops = (four) will be on the display of each of the monitors, each cluster is the entire display can not be cut - part, mother The clusters then perform a decompression function on the received transitions, and then the decompressed data is displayed by each pixel of each cluster. 8. If you apply for Meizhan Ren-Shi, the towel display device can display the general display information. 9. If the application is to continue the high-resolution display described in the above-mentioned item, the display is displayed to display the holographic display data. A high-resolution display according to any of the preceding claims, wherein the space for performing the compression calculation is located on the same substrate as the pixels of the display. 11. A south resolution display as described in the first to fourth pages of the patent application, wherein the space used to perform the compression calculation is different from the pixels of the display - 158 200916986 12. As claimed in the patent range 7 to u The high-resolution display of any one of the preceding claims, wherein the copper and vertical ship receiving data for calculating the ageing axis 4 are calculated. 13. The high-resolution display device according to any one of the seventh and the fourth aspect of the invention, wherein each of the clusters for decompressing &amp; 3 Boss receives data in a row by means of a concurrent material. The high-resolution display described in any of the above-mentioned patents No. 7 ^JU, wherein each of the plexus axes for the decompression calculation - the bribe connection receiving information. The high-resolution display of the nin-item is a high-resolution display. 16. As for the patent application, the 7th compression will be performed by each of the high-resolution displays of any of 40 to 15 The solution is executed within ms or less. 17. The 9th hologram image calculation as claimed in the patent application will be performed after decompressing to the high resolution display of any of the 16 items. '18 of which apply for a patent A computational display performed by the encoding of the high-resolution inter-display optical modulation H described in the first item, wherein the determination is made by using the same substrate disposed on each pixel 159 200916986 of the spatial light modulator. The circuit components on the implementation. 19. The high-resolution display of claim 18, wherein the calculation performed to determine the encoding of the spatial light modulator does not involve Fourier transform ((10) plus ffooO or Philippine conversion (such as coffee bans) &lt; 20. The high-resolution display of claim 18 or 19, wherein the space is used to negotiate a space between the four elements of the thief (four). A high-resolution display according to any one of claims 18 to 12, wherein the calculation performed by the coding of ^=: spatial light tones (four) is dispersed. The high-resolution display includes a thin-film transistor as described in any one of the preceding claims, wherein at least one of the circuit component packages is included in the field in the display. Part 23. The effective location of the circuit component of the high-resolution display of any of the preceding claims, wherein the high-resolution display is as described in the patent application. among them The effective area of the circuit component is composed of a continuous grain. The high-resolution display of any of the foregoing items, wherein at least a part of the circuit components have a valid effective area 26. A high-resolution display according to any of the preceding claims, wherein at least a portion of the circuit component has an effective location consisting of a single crystal germanium. The high-resolution display of the above-mentioned item, wherein the effective area of at least a part of the circuit component is composed of a single crystal grain. 28. The high-resolution display according to any one of the preceding claims, At least some of the circuit components have effective locations formed by organic semiconductors. 29. The high resolution display of any of the preceding claims, wherein the substrate is monocrystalline. 30. The high resolution display of claim 19, wherein the display is fabricated using a twin crystal liquid crystal technology. The high-resolution display of any one of claims 1 to 28, wherein the substrate is made of glass. 32. The high resolution display of any of the preceding claims, wherein the video frame rate is at least about 25 Hz. 33. The high resolution display of any of claims 9 to 32, wherein only real-world image data is transmitted to the display. 34. The high-resolution display of claim 33, wherein the real-world image data is composed of light intensity and depth correspondence data. The high-resolution display of any one of claims 17 to 34, wherein the hologram is calculated as an immediate or near real-time operation. The high-resolution display of any one of claims 17 to 35, wherein the holographic calculation performed is performed using a view table method. 37. The high resolution display of any of claims 17 to 36, wherein the sub-image is used for holographic calculation. 162. The high-resolution display of any one of claims 1 to 29, wherein the display employs MEMS technology in manufacturing. 39. The high resolution display of any of claims 1 to 29, wherein the display is to employ field emission display (FED) technology. 40. A method of displaying a decompressed high resolution image material using the high resolution display of any of the preceding patent claims, comprising the steps of: (a) compressing high resolution image data; (b) transmitting through Compressed high-resolution imagery; (c) receiving compressed high-resolution imagery; (d) decompressing compressed high-resolution imagery; and (e) decompressing high-resolution images The data is displayed on the display. 163
TW097118290A 2007-05-16 2008-05-16 A high-resolution display of highly decompressed image data can be displayed TWI464548B (en)

Applications Claiming Priority (5)

Application Number Priority Date Filing Date Title
GB0709379A GB0709379D0 (en) 2007-05-16 2007-05-16 Smart display extended
GBGB0718602.6A GB0718602D0 (en) 2007-05-16 2007-09-25 Holograms
GBGB0718654.7A GB0718654D0 (en) 2007-05-16 2007-09-25 Holograms
GBGB0718656.2A GB0718656D0 (en) 2007-05-16 2007-09-25 Holograms
GBGB0718619.0A GB0718619D0 (en) 2007-05-16 2007-09-25 Holograms

Publications (2)

Publication Number Publication Date
TW200916986A true TW200916986A (en) 2009-04-16
TWI464548B TWI464548B (en) 2014-12-11

Family

ID=38234523

Family Applications (3)

Application Number Title Priority Date Filing Date
TW097118290A TWI464548B (en) 2007-05-16 2008-05-16 A high-resolution display of highly decompressed image data can be displayed
TW097118291A TWI447540B (en) 2007-05-16 2008-05-16 Holographic display with communications
TW97118288A TWI412904B (en) 2007-05-16 2008-05-16 A holographic display and method of generating a holographic reconstruction of a three dimensional scene

Family Applications After (2)

Application Number Title Priority Date Filing Date
TW097118291A TWI447540B (en) 2007-05-16 2008-05-16 Holographic display with communications
TW97118288A TWI412904B (en) 2007-05-16 2008-05-16 A holographic display and method of generating a holographic reconstruction of a three dimensional scene

Country Status (2)

Country Link
GB (1) GB0709379D0 (en)
TW (3) TWI464548B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9892669B2 (en) 2014-03-18 2018-02-13 Nvidia Corporation Superresolution display using cascaded panels

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI500313B (en) * 2010-06-04 2015-09-11 Au Optronics Corp Image scanning system and method of controlling the same
CN107340704B (en) * 2017-01-04 2020-02-07 京东方科技集团股份有限公司 Holographic display device
CA3101587C (en) * 2018-06-29 2023-09-19 Leia Inc. Mixed-format backlight, display, and method

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH05119341A (en) * 1991-10-25 1993-05-18 Seiko Epson Corp Optical device
US6122403A (en) * 1995-07-27 2000-09-19 Digimarc Corporation Computer system linked by using information in data objects
US6438266B1 (en) * 1998-08-27 2002-08-20 Lucent Technologies Inc. Encoding images of 3-D objects with improved rendering time and transmission processes
GB2350962A (en) * 1999-06-09 2000-12-13 Secr Defence Brit Holographic displays
JP2001285894A (en) * 2000-03-31 2001-10-12 Olympus Optical Co Ltd Method for running three-dimensional image data
JP2002123160A (en) * 2000-10-16 2002-04-26 Sony Corp Holographic stereogram printing order accepting system and method
US6582980B2 (en) * 2001-01-30 2003-06-24 Eastman Kodak Company System for integrating digital control with common substrate display devices
JP3847593B2 (en) * 2001-02-20 2006-11-22 シャープ株式会社 Optical element such as corner cube array and reflective display device including the same
US7272575B2 (en) * 2001-07-13 2007-09-18 Lilly Mae Vega Method and system for facilitating service transactions
KR100891293B1 (en) * 2002-11-13 2009-04-06 씨리얼 테크놀로지스 게엠베하 Method and device for reconstructing a three-dimensional hologram, and hologram-bearing medium
JP4464751B2 (en) * 2003-08-25 2010-05-19 株式会社エヌ・ティ・ティ・ドコモ Stereoscopic image display apparatus and stereoscopic image display system
DE102004063838A1 (en) * 2004-12-23 2006-07-06 Seereal Technologies Gmbh Method and apparatus for calculating computer generated video holograms
TWI268737B (en) * 2005-07-29 2006-12-11 Au Optronics Corp Full-color organic electroluminescence panel with high resolution
TWM288379U (en) * 2005-08-26 2006-03-01 Jeng Jang Three-dimensional, omni-directional holographic display
US8903916B2 (en) * 2006-07-05 2014-12-02 International Business Machines Corporation Method, system, and computer-readable medium to render repeatable data objects streamed over a network
US7702164B2 (en) * 2006-09-22 2010-04-20 International Business Machines Corporation Run length limited encoding of data into a 5×5 matrix for recording into a holographic medium

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9892669B2 (en) 2014-03-18 2018-02-13 Nvidia Corporation Superresolution display using cascaded panels
US9934714B2 (en) 2014-03-18 2018-04-03 Nvidia Corporation Superresolution display using cascaded panels

Also Published As

Publication number Publication date
TWI464548B (en) 2014-12-11
TW200928626A (en) 2009-07-01
GB0709379D0 (en) 2007-06-27
TWI447540B (en) 2014-08-01
TW200916985A (en) 2009-04-16
TWI412904B (en) 2013-10-21

Similar Documents

Publication Publication Date Title
US11269295B2 (en) Holographic display
TW200912888A (en) Holographic display
CN101743519B (en) Holographic display
US8218211B2 (en) Holographic display with a variable beam deflection
TW200916986A (en) High resolution display
Thatte Cinematic virtual reality with head-motion parallax