516005 #ir^，ii& 變 K、1rf内容是^/1|..‘予修31〇 五、發明說明（1) 本發明係關電腦產生的全像影像。 全像術是一種產生3 D影像具有全部深度暗示存在的技 術。電腦產生全像圖（CGH)是一種形成不存在的（合成）物 件影像的重要方法。假設合成影像可藉某種形式的資料結 構說明，然後，電腦可算出某個設計平面的全像圖樣。此 點相當於習知（干涉計量法）全像圖形成時藉光敏材料記錄 干涉圖樣。然後，算出的CGH圖樣應用至空間光調變器 (SLM)，其調變入射其上被讀出的光。此種調變後的光傳 播而獲得預定3D影像，可由觀視者觀視。 有關全像顯示器的一項考量是放大影像大小I和關聯的 視野F0V。對一指定CGH而言，由亮度理論可預期增加I ，通常導致F0V對應減小。爲了獲得良好估値 I.F0V 常數 可加大此種乘積値的技術相當重要，原因在於一成不變 地，I及/或FOV的限制決定了 CGH處理的複雜程度，而 非影像解像度。 CGH SLM顯示器特有的解像度限制，要求CGH繞射的光 藉照比例縮放的光學裝置放大，而產生具有所需大小I和 視野FOV(FOV是一種量，定義觀視者眼睛必須放置來看全 部由全像圖產生的影像的位置區域）的影像。若採用習知 辦法，此等光學裝置一成不變地將失真和光像差導入縮放 影像。用於高階用途，此種影像品質的低劣無法接受。 先前試圖校正因縮放光學裝置導入的光像差，包括相位 516005 修麵 正詳 革 ^ 〇, f γΙ nt\' f< Λ:516005 # ir ^, ii & change K, 1rf content is ^ / 1 | .. ‘Yiu Xiu 31〇 V. Description of the invention (1) The present invention relates to a holographic image generated by a computer. Holography is a technique that produces 3D images with full depth suggestion of existence. Computer-generated holograms (CGH) are an important method for forming images of non-existent (synthetic) objects. Suppose that the synthetic image can be explained by some form of data structure, and then the computer can calculate a hologram of a design plane. This point is equivalent to recording the interference pattern by photosensitive material when the conventional (interferometry) hologram is formed. The calculated CGH pattern is then applied to a spatial light modulator (SLM), which modulates the light incident on it to be read out. This modulated light is transmitted to obtain a predetermined 3D image, which can be viewed by a viewer. One consideration with holographic displays is the enlarged image size I and the associated field of view F0V. For a given CGH, an increase in I can be expected from the brightness theory, which usually results in a corresponding decrease in F0V. In order to obtain a good estimate of the I.F0V constant, it is important to increase the product of this technique, because the I and / or FOV constraints invariably determine the complexity of the CGH process, not the image resolution. CGH SLM displays have unique resolution limitations that require CGH diffracted light to be magnified by an optical device that scales to produce a desired size I and field of view FOV (FOV is a quantity that defines the viewer ’s eyes must be placed to see all (The location area of the image generated by the hologram). Using conventional methods, these optical devices consistently introduce distortion and light aberrations into the scaled image. For high-end applications, this image quality is unacceptable. Previous attempts to correct the optical aberrations introduced by the zoom optics, including phase 516005 Shaving Correct details ^ 〇, f γΙ nt \ 'f < Λ:
iE # 。-1 五、發明說明（2) 屏障倍增器構想，例如述於 Maeno K， Fukaya N, Nishikawa 0, Sato Κ, Honda Τ, 「使用 15 百萬像素 LCD之電氣全像顯示器」，Proc SPIE，2652期，15-23頁 (1 996年）。相位屏障倍增器涉及藉二維相位分散倍增全像 圖，該二維相位分散係與正在被重新組構的影像無關。不 幸，此種技術仰賴傍軸估算，因此具有有意義的F0V的大 影像校正的成功有限。 根據本發明提供一種產生電腦產生全像圖（CGH)資料之 方法，用以應用至一全像顯示器的空間光調變器而顯示三 維全像影像，該方法包含： 決定於全像顯示器的光學組件之光像差；以及 界定該CGH資料俾補償測得的光像差的影響。 CGH可設計完成或部分補償光學組件的影響。可產生應 用至SLM的圖樣，故由顯示器形成完美或近完美的全像影 像。 光學組件引起的光像差的決定可藉多種技術進行，但較 佳包含決定由光學組件導入的光線徑長。 因此介於空間光調變器與三維全像影像間的光線光徑長 度可用以定義CGH資料。 CGH資料可定義成··由空間調變器射出的光，可補償由 光學組件導入的光像差。 爲了讓電腦記憶體的需求減至最低，而未導入過多運算 時間額外負擔，較佳使用內插功能來決定介於空間光調變 516005 五、發明說明（3) 器與三維全像影像間的光線光徑長度。 另外或此外，決定由顯示器的光學組件導入的光像差所 得結果可儲存在查表。此點特別可用於需產生重新配置影 像或動態影像的應用用途。 光學系統可設計成具有特殊光像差，故當於定義CGH資 料中進行光像差的補償時，也可達成增加三維全像的視野 码η,δ'-Γ·:〈 ί 年万为丨 u^’M-t ^ixc^ife9?ul::、1:f.tv、^J--iM.1r‘#iE。 光學組件包含一或多面曲面鏡。此種情況下，光像差效 應的補償執行方式，可藉由基於觀視者的觀視角度，決定 由空間光調變器產生的一影像位置和方向性，因而產生一 個I三維全像影像位在凸面鏡後方，該影像於觀視者移動時 穩定出現。 根據本發明之第二特徵方面，提供一種全像顯示器，包 含·· 一電腦，設置用以產生及/或儲存電腦產生全像圖（CGH) 資料； 一空間光調變器，耦合至電腦用以接收CGH資料； 光學組件，設置於空間光調變器的光徑上，用以造成欲 顯示的三維全像影像； 其中CGH資料係定義成可補償由光學組件引起的光像差 效應。 根據本發明之第三特徵方面，提供一種電腦儲存媒體， 其上儲存意圖於全像顯示器的空間光調變器形成一光調變 516005 五、發明說明（4) 圖樣用的電腦產生全像圖（CGH)資料，該CGH資料包括對 顯示器的光學組件發生的光像差效應的補償。 須了解使用「光像差」一詞意欲指可能造成顯示的影像 偏離意欲的全像影像的失真、旋轉、平移等。光像差可能 來自於使用完美或非完美光學組件。 圖式之簡單說明 爲求更了解本發明且爲了顯示如何實施，現在參照附圖 舉例說明本發明，附圖者： 圖1舉例說明產生一電腦產生全像圖之相干性光線追蹤 方法； 圖2A舉例解明於偏心軸拋物面鏡系統的三個直排點之 不完美單色成像； 圖2B舉例說明圖2A之三點之不完美成像的進一步細節 ， 圖3顯示校正後的CGH，產生4點直排繞射限制點； 圖4舉例說明由影像上各點至各CGH抽樣點之光徑計算 9 圖5顯示一影像的三個不同視圖；以及 圖6顯示以曲面鏡所得觀視角度的加大。 發明之詳細說明 全像圖產生的基本工作是產生一圖樣，其將繞射光線而 產生所需3D影像，結知爲達於此項目的，首先一物件以 相干性照明。來自物件的光傳播至一設計平面。由該設計 516005 修讀 11 疋外,f ί f* ：I 緣叫 it揭 ° -t: 五、發明說明（5) 平面的光域知識，可產生一顛倒光域，該顛倒光域當形成 時將允許於空間再生所需3D影像。全像術是一種讓此現 象變成可能的技術，原因在於全像術儲存且允許重新播放 複雜的光域圖樣於設計平面。 電腦產生全像圖（CGH)允許此種處理對合成（不存在）物 件進行。決定所需CGH圖樣的第一階段係計算由合成物件 產生的電場分布。欲外加至空間光調變器（SLM)的所需圖 樣係由於設計面的此種電場分布決定。 圖1顯示一種用以於設計平面，決定典型CGH外加至 SLM之圖樣之技術。對虛擬物件2相對於設計平面4於空 間的座標產生一資料域。設計平面被像素化，而光線由物 件上的該像素「可見的」各點被相干性追蹤至各像素。由 各光線引起的複雜電場於各像素被相干性加總。可結合複 雜化的放射性模型來準確決定光線強度（來自物件表面、 多重光源和反射的鏡面反射和漫射反射）。阻隔效應自動 被包括在內，原因在於唯有來自於一特定像素「可見的」 物件部分的光線才被「追蹤」，且促成該像素的電場強度 。此項技術稱作相干性光追蹤技術（CRT)。 已經計算於設計平面的複雜電場，此項資料在被光波重 播而產生所需3D影像前，須被儲存於某種模組結構。全 像攝影技術，其構想爲參考波，允許複雜的電場0分布被 儲存爲實量。如此允許使用振幅或相位調變系統來產生所 需3D影像。 516005 五、發明說明（6 ) 爲了產生所需大小的影像，通常設置縮放光學裝置。圖 2A顯示CGH6之簡單光線追蹤，CGH以尋常方式設計，使 用偏心軸拋物面鏡1 2產生三個直排點8，三點與觀視者 1 〇眼睛等距。光線追蹤分析（細節顯示於圖2B )顯示通常 重新組構模糊點。雖然中點1 4淸晰鮮明且在預定位置， 但兩邊的兩點1 6，1 8皆模糊且不再於要求的平面一光線20 不會全部交叉在空間的要求點，或確實位在任何單一位置iE #. -1 V. Description of the invention (2) The concept of barrier multiplier, for example, described in Maeno K, Fukaya N, Nishikawa 0, Sato K, Honda T, "Electric Holographic Display Using 15 Megapixel LCD", Proc SPIE, 2652 Issue, pp. 15-23 (1,996). The phase barrier multiplier involves multiplying the hologram by two-dimensional phase dispersion, which is independent of the image being restructured. Unfortunately, this technique relies on paraxial estimation, so large image corrections with meaningful F0V have limited success. According to the present invention, there is provided a method for generating computer-generated hologram (CGH) data for displaying a three-dimensional holographic image using a spatial light modulator of a holographic display. The method includes: The optical aberration of the component; and defining the CGH data to compensate for the effect of the measured optical aberration. CGH can be designed to complete or partially compensate for the effects of optical components. It can produce patterns applied to SLM, so the monitor forms perfect or near-perfect holographic images. The determination of optical aberrations caused by optical components can be made by a variety of techniques, but it is better to include the determination of the length of the light rays introduced by the optical components. Therefore, the light path length between the spatial light modulator and the 3D holographic image can be used to define the CGH data. CGH data can be defined as the light emitted by the spatial modulator, which can compensate for the optical aberrations introduced by the optical components. In order to minimize the computer memory requirements without introducing extra burden of computing time, it is better to use interpolation to determine the spatial light modulation. 516005 V. Description of the invention (3) between the device and the 3D holographic image The light path length. In addition or in addition, the results obtained by determining the optical aberrations introduced by the optical components of the display may be stored in a lookup table. This is particularly useful for applications where reconfiguration or motion images are required. The optical system can be designed to have special optical aberrations, so when compensating for optical aberrations in the definition of CGH data, it is also possible to achieve an increase in the field code η, δ'-Γ · of the three-dimensional full image: u ^ 'Mt ^ ixc ^ ife9? ul ::, 1: f.tv, ^ J--iM.1r' # iE. The optical assembly includes one or more curved mirrors. In this case, the compensation implementation of the optical aberration effect can determine the position and directionality of an image generated by the spatial light modulator based on the viewing angle of the viewer, thereby generating an I 3D hologram Located behind the convex mirror, the image appears steadily as the viewer moves. According to a second characteristic aspect of the present invention, there is provided a holographic display, comprising a computer configured to generate and / or store computer generated hologram (CGH) data; a spatial light modulator coupled to a computer The CGH data is received; the optical component is set on the optical path of the spatial light modulator to create a three-dimensional holographic image to be displayed; the CGH data is defined to compensate the optical aberration effect caused by the optical component. According to a third characteristic aspect of the present invention, there is provided a computer storage medium on which a spatial light modulator intended for a holographic display is formed to form a light modulation 516005 V. Description of the invention (4) A computer for drawing generates a holographic image (CGH) data, which includes compensation for optical aberration effects occurring in the optical components of the display. It is important to understand that the use of the term "light aberration" is intended to mean distortion, rotation, translation, etc. that may cause the displayed image to deviate from the intended holographic image. Optical aberrations may result from using perfect or imperfect optical components. Brief description of the drawings In order to better understand the present invention and to show how to implement it, the present invention will now be described by way of example with reference to the accompanying drawings, which are: Figure 1 illustrates a coherent ray tracing method for generating a computer-generated hologram; Figure 2A An example illustrates the imperfect monochrome imaging of the three in-line points of the eccentric parabolic mirror system; Figure 2B illustrates further details of the imperfect imaging of the three points of Figure 2A. Figure 3 shows the corrected CGH, resulting in a 4-point straight Row diffraction limit points; Figure 4 illustrates the calculation of the light path from each point on the image to each CGH sampling point 9 Figure 5 shows three different views of an image; and Figure 6 shows an increase in viewing angle obtained with a curved mirror . Detailed description of the invention The basic work of hologram generation is to generate a pattern that will diffract light to produce the required 3D image. It is known that for this project, an object is first illuminated with coherence. Light from the object travels to a design plane. From this design 516005, read 11 疋 外, f ί f *: I edge is called it expose ° -t: V. Description of the invention (5) The knowledge of the plane light domain can produce an inverted light domain, which should form It will allow the 3D images required for spatial reproduction. Holography is a technique that makes this phenomenon possible because holography stores and allows the complex light domain patterns to be replayed on the design plane. Computer generated holograms (CGH) allow this process to be performed on synthetic (non-existent) objects. The first stage in determining the required CGH pattern is to calculate the electric field distribution generated by the composite object. The required pattern to be added to the space light modulator (SLM) is determined by this electric field distribution on the design surface. Figure 1 shows a technique used in the design plane to determine a typical CGH pattern added to the SLM. A data field is generated for the coordinates of the virtual object 2 with respect to the design plane 4 in space. The design plane is pixelated, and the light is “coherently” tracked from pixel to point by the pixel on the object. The complex electric field caused by each light is coherently summed at each pixel. Complicated radioactivity models can be combined to accurately determine light intensity (specular and diffuse reflections from object surfaces, multiple light sources, and reflections). The blocking effect is automatically included because only the light from the "visible" part of a particular pixel is "traced" and contributes to the pixel's electric field strength. This technique is called coherent light tracking (CRT). The complex electric field that has been calculated on the design plane must be stored in a certain module structure before being reproduced by light waves to produce the required 3D image. Holographic imaging technology is conceived as a reference wave, allowing complex electric field 0 distributions to be stored as real quantities. This allows the use of amplitude or phase modulation systems to produce the required 3D images. 516005 V. Description of the invention (6) In order to produce an image of a desired size, a zoom optical device is usually provided. Figure 2A shows the simple ray tracing of CGH6. CGH is designed in the usual way, using an eccentric axis parabolic mirror 12 to generate three straight points 8, which are equidistant from the viewer's 10 eyes. Ray tracing analysis (details shown in Figure 2B) show that blur points are usually restructured. Although the midpoint 14 is clear and in a predetermined position, the two points 16 and 18 on both sides are blurred and no longer on the required plane. A ray 20 will not all cross at the required point in space, or indeed be located at any Single location
〇之 。這三點的影像16, 18對觀視者而言，顯然位在不正確位 置且模糊，而其表觀位置將隨觀視者所在位置遷移。 只要CGH經妥爲設計，即可校正此等效應。校正技術可 視爲反向光線追蹤。對所示系統而言，來自物件各點 14, 16, 18至CGH設計平面6上的各抽樣點的光線20路徑 已經算出。光徑長度資訊決定位在設計平面6的校正後的 CGH圖樣，因而若CGH經組構與重播，則三點影像將適當 產生作爲正確（亦即設計）所在位置的繞射限制點。當如前 述發生複雜光域的相干性加總時，進行於設計平面6因不 同光徑長度造成的CGH圖樣校正。換言之，當計算CGH圖 樣應用至SLM時，使用光離開SLM後，光通過光學裝置的 表現。 圖3顯示此種系統在操作中。實例顯示有四個直排物件 點22, 24,26, 28，且全部皆正確界定於同一平面，亦即不 模糊或漂移。 雖然前文係簡單例，但該項技術原則上可透過多種不完 516005 事 五、發明說明（7) 美的光學元件來產生接近任意的3D影像，規定光學元件 的特性爲原先已知。補償多種光學元件的能力相當重要。 舉例言之，使用球形光學形式（與上圖使用非球形相反）允 許節省成本，此點對偶爾因系統含有高度複雜的CGH而需 要的大孔徑光學裝置特別顯著。 一項重要考量是前述辦法的運算需求。對有顯著容積通 過元件系統的影像，決定所需光徑絕非不起眼的工作。即 使對圖2顯示的單元件系統，也無法決定光徑的一般分析 '之道一典型須使用Fermat’s氏原理的數値解。 用於需要產生可重新配置或動態影像的應用用途，可能 需要查表辦法（基於對感興趣的光學系統預先算出的資訊） ，而非即時計算光徑長度。經由注意到光徑長度典型通過 物件空間和CGH設計平面而徐徐改變，故可能進一步簡化 。因此可相當準確的在少數間隔均勻的「抽樣」點即可相 當準確算出光徑長度，然後基於此等數値匹配一曲線，而 允許內插樣本値以外的各點。因此可使用簡單內插功能而 減少電腦記憶體需求，且不會導入過多運算時間額外開銷 〇 圖4顯示對一特定光學元件形狀4而言，由具有座標 Up，yp，zp}的影像上一點30,至具有座標Ue，ye，ze09CGH 抽樣點32光徑計算之代表圖。此需數値解。圖中’假設 屬鏡面，光徑計算要求決定光線撞及鏡面的位置34 ’座標 是 Um，ym，。 516005 正讀 $赛 ψί g〇 之. These three points of images 16, 18 are apparently to the viewer in an incorrect position and blurred, and their apparent position will shift with the viewer's position. As long as CGH is properly designed, these effects can be corrected. The correction technique can be viewed as reverse ray tracing. For the system shown, the path of light 20 from each point 14, 16, 18 of the object to each sampling point on the CGH design plane 6 has been calculated. The optical path length information determines the corrected CGH pattern located on the design plane 6. Therefore, if the CGH is structured and replayed, the three-point image will be appropriately generated as the diffraction limit point at the correct (ie, designed) location. When the coherence summation of the complex light domain occurs as described above, the CGH pattern correction on the design plane 6 due to different optical path lengths is performed. In other words, when the CGH pattern is calculated and applied to the SLM, the performance of the light passing through the optical device after the light leaves the SLM is used. Figure 3 shows such a system in operation. The example shows that there are four in-line object points 22, 24, 26, and 28, all of which are correctly defined on the same plane, that is, not blurred or drifted. Although the foregoing is a simple example, this technology can in principle use a variety of endless 516005 events. 5. Description of the Invention (7) Beautiful optical elements can generate nearly arbitrary 3D images, which stipulate that the characteristics of optical elements are previously known. The ability to compensate for a variety of optical components is important. For example, the use of spherical optics (as opposed to the use of asphericity in the figure above) allows cost savings, which is particularly significant for large-aperture optics that are occasionally required because the system contains highly complex CGH. An important consideration is the computational requirements of the previous approach. For images with a significant volume passing through the component system, determining the required light path is not a trivial task. Even for the single-element system shown in Fig. 2, the general analysis of the optical path cannot be determined. The way-a typical solution using Fermat's principle is required. For applications that require reconfigurable or dynamic images, a table lookup (based on pre-calculated information for the optical system of interest) may be required instead of calculating the optical path length in real time. It is possible to simplify further by noting that the optical path length typically changes slowly through the object space and the CGH design plane. Therefore, it is possible to relatively accurately calculate the optical path length from a few evenly spaced "sampling" points, and then match a curve based on these numbers to allow interpolation of points other than the sample. Therefore, the simple interpolation function can be used to reduce the computer memory requirements without introducing excessive computing time overhead. Figure 4 shows that for a specific optical element shape 4, the image with the coordinates Up, yp, zp} 30. Representative diagram for calculation of 32 optical paths to sampling points with coordinates Ue, ye, ze09CGH. This requires some explanation. In the figure, assuming that it is a mirror surface, the calculation of the optical path requires determining the position where the light hits the mirror surface. 34 'The coordinates are Um, ym ,. 516005 Reading $ 赛 ψί g
五、發明說明（8) CGH可被想成根據觀視者的視角’提供不同影像。典型 此等影像爲對實際物件預期的適當透視圖，但非必然如此 。用於觀視者觀視CGH產生的立方體影像之例’ CGH重播 特有的有限範圍觀視角，表示觀視者通常只能看到環繞立 方體側邊的一小段。藉由設計CGH ’使重播於觀視角偏離 軸線移動時顯示旋轉，觀視者可進一步「看到」各邊。 圖5驗證此項技術，顯示如何當觀視角改變時，藉由旋 轉影像，讓觀視者可進一步觀視立方體整圈的影像。圖5A 中，觀視者3 6於立方體的其中一面，接近筆直觀視立方 體40影像，亦即實質上沿軸線3 8觀視。圖5B中，觀視 者36移離軸線38而加大觀視角，且立方體40已經於反 向旋轉，讓他可見更多頂面。圖5C中，觀視者36移動更 遠，立方體40影像旋轉，故他幾乎直接注視頂面。如此 當然表示影像已經不再像個真實物件的表現（可能是非期 望的），且需識別光像差或非成像光學裝置，允許影像穩 定呈現，同時允許寬廣觀視角。 初步使用簡單光學裝置考量，指出這種構想並非不合理 。經由「鏡廳」效應的啓示（其說明參照圖6 )，可想像 CGH影像上各點僅「發光」至相當小角度的角錐（結果導致 該點只可由極小範圍的角度可見）。若光線射在凸面鏡上 ，則可見反射時，光線散出，而有效加大觀視角。現在， 當於此種曲面鏡觀視物件時，影像顯然隨著觀視者的觀視 角旋轉。可在CGH設計時有效取消此種旋轉。但也顯然物 -10- 5160055. Description of the invention (8) CGH can be thought of as providing different images depending on the viewer's perspective. These images are typically appropriate perspective views of actual objects, but not necessarily so. An example of a cube image produced by a viewer watching CGH ’CGH Replay A unique limited range viewing angle, which means that the viewer usually only sees a small section around the sides of the cube. By designing the CGH 'to replay the display rotation when the viewing angle moves away from the axis, the viewer can further "see" the sides. Figure 5 validates this technique, showing how the viewer can further view the entire image of the cube by rotating the image when the viewing angle is changed. In FIG. 5A, the viewer 36 is on one side of the cube, and is close to the pen to visualize the cube 40 image, that is, viewed substantially along the axis 38. In FIG. 5B, the viewer 36 moves away from the axis 38 to increase the viewing angle, and the cube 40 has been rotated in the opposite direction, so that he can see more of the top surface. In FIG. 5C, the viewer 36 moves farther and the image of the cube 40 rotates, so he looks at the top surface almost directly. This, of course, means that the image no longer behaves like a real object (may be undesired), and it is necessary to identify optical aberrations or non-imaging optical devices to allow the image to be presented stably and at the same time allow a wide viewing angle. Preliminary considerations of using simple optical devices point out that this concept is not unreasonable. Through the enlightenment of the "mirror hall" effect (refer to Figure 6 for its explanation), it is conceivable that each point on the CGH image only "glows" to a fairly small angle pyramid (as a result, the point can only be seen from a very small range of angles). If the light hits the convex mirror, the light will diffuse when visible reflection, effectively increasing the viewing angle. Now when viewing objects with such curved mirrors, the image obviously rotates with the viewing angle of the viewer. This rotation can be effectively cancelled during CGH design. But also obviously -10- 516005
° > 五、發明說明（9) 件隨觀視角移動。此點可經補償，但CGH重播時可利用的 影像容積有限，將限制可使用的位置範圍，因而限制影像 的大小。 當觀視者42沿凸面鏡44軸線48觀視系統時，由CGH( 圖中未顯示）產生的影像46係位在軸線48上，故反射50 ( 因而由觀視者所見影像）係位在鏡44後方軸線上。當觀視 者52移離軸線，故由不同角度注視時，CGH的相位資訊可 安排成CGH產生的影像46被移動與旋轉至新位置54。換 言之，SLM編碼資料使用有關光線離開SLM後的表現資訊 ，來提供當觀視者注視CGH不同部分時的不同影像。當偏 心軸觀視者5 2觀視系統時，影像5 4的反射5 0將出現在 系統沿軸線觀視時的相同位置。但相對於影像軸線，繞射 角仍小。需要分束鏡（圖中未顯示）來允許觀視反射影像， 不被CGH顯示系統遮蔽。 業界人士須了解本發明非僅限於前述具體實施例。例如 已經說明藉相干性光線追蹤技術（CRT )計算CGH。但是發明 也可用於藉其它任一種技術計算的CGH。即使最終CGH非 藉CRT方法設計，相干性光線追蹤仍可用以決定需要的光 像差校正。光像差補償也可藉由查驗真實系統中由CGH產 生的測試影像，迭代變更CGH，直至產生需要的影像爲止 而達成。像素追蹤像素像素° > V. Description of the invention (9) The item moves with the viewing angle. This point can be compensated, but the image volume available during CGH replay is limited, which will limit the range of positions that can be used, and therefore the size of the image. When the viewer 42 views the system along the axis 48 of the convex mirror 44, the image 46 generated by the CGH (not shown) is located on the axis 48, so the reflection 50 (thus the image seen by the viewer) is located in the mirror 44 on the rear axis. When the viewer 52 moves away from the axis, and therefore looks from different angles, the phase information of the CGH can be arranged such that the image 46 generated by the CGH is moved and rotated to a new position 54. In other words, SLM-encoded data uses performance information about light leaving the SLM to provide different images when the viewer looks at different parts of the CGH. When the eccentric axis viewer 52 views the system, the reflection 50 of the image 54 will appear at the same position when the system views along the axis. However, the diffraction angle is still small relative to the image axis. Beamsplitters (not shown) are required to allow viewing of reflected images without being obscured by the CGH display system. Those skilled in the art must understand that the present invention is not limited to the foregoing specific embodiments. For example, it has been described that CGH is calculated by coherent ray tracing technology (CRT). But the invention can also be used to calculate CGH by any other technique. Even if the final CGH is not designed by the CRT method, coherent ray tracing can still be used to determine the required optical aberration correction. Optical aberration compensation can also be achieved by examining the test images generated by CGH in a real system and iteratively changing the CGH until the required image is produced. Pixel tracking pixel pixel
-11- 516005-11- 516005
IIII
五、發明說明（10 元件符號說明 2 4 6 8 10 12 14 16，18 20 22,24,26,28 30 32 36 38 40 42 44 46 48 50 52 54 CRT F〇V CGH GLM 虛擬物件 設計平面 CGH(電腦產生全像圖） 直排點 觀視者 偏心軸拋物面鏡 中點（物件點） 點（物件點） 光線 直排物件點 點 抽樣點 觀視者 軸線 立方體 觀視者 凸面鏡 影像 軸線 反射 觀視者 新位置 相干性光追蹤技術 視野 電腦產生全像圖 空間光調變器 -12-V. Description of the invention (10 Description of component symbols 2 4 6 8 10 12 14 16, 18 20 22, 24, 26, 28 30 32 36 38 40 42 44 46 48 50 52 54 CRT F0V CGH GLM Virtual object design plane CGH (Computer generated hologram) Straight point viewers Eccentric axis parabolic mirror Midpoint (object point) Point (object point) Rays straight object points Sample points Viewer axis Cube viewer Convex mirror image axis Reflection view New position coherent light tracking technology field of vision computer generated hologram spatial light modulator-12-