1289024 九、發明說明： 【發明所屬之技術領域】 - 本發明係有關於視訊播放裝置，尤指標誌（logo)處理方法與 、 電路。 【先前技術】 標誌（logo)係廣泛地被應用於電視節目中，用來宣告該些電 • 視節目之提供者或製作者的版權。一光碟片，例如數位多用途光 碟片（DigitalVersatileD—DVD)中所記錄之一節目裡也可能有 類似用途之標誌。 由於節目中之標誌、並非直接與節目β容相_，且該標誌有時會 部分地擋住節目中所顯示之特定視訊物件（vide〇Gb⑽所以大 多數觀看節目之觀眾對於佔據電視機或顯示器的榮幕/面板之一 # 祕_域麻煩。糾，若該縣於大部分咖佔據著同一 角洛，尤其是對於那些採用電聚顯示面板㉞纖⑽㈣一， 之電職或顯示器而言’日後該電漿顯示面板上絲顯示該 厂之顯不早凡很容易燒毁或者性能相較於其它顯示單元變得較 除rsr電視節目中之一第一標諸可能被去 二Γ這是因為該第一標諸與節目内容自服務 R者無線或树地傳輸轉觀財讀，絲被混合在一 1289024 起。然而，若該數位電視節目巾有—第二標諸縣就被混合於節 目内各’則依據上述該些規範之標認去除架構對於該第二標錄就 永遠無效。 【發明内容】 本發明之目的之-在於提供標誌、（lGgG)處理方法與電路，以 偵測原本就被混合於節目内容之至少一標誌。 本發明之另一目的在於提供標誌處理方法與電路，以改變原本 就被混合於節目内容之至少一標誌之至少一色彩。 依據本發明-實關，係揭露-種標域理方法。該標諸處理 方法包含有於依序接收自一視訊來源（videosource)之複數個視 訊圖框（vide0frame)中，比較一基準圖框（baseftame)與其它 視訊圖框的至少-部分以產生—遮罩（mask)，該遮罩具有可能對 應於至少一標誌之至少一候選區域。該標誌處理方法另包含有對 該遮罩進行後處理（P〇stpr〇CeSSing)以修正該至少一候選區域， 以及將該至少一候選區域中之至少一者選擇為標誌區域以供指出 (indicate)該至少一標誌。 依據本發明一實施例，係揭露一種標誌處理方法。該標誌處理 方法包含有偵測至少一標誌，以及改變該至少_標誌之至少一色 1289024 依據本發明-實施例，係揭露—種麟處理電路。該標該處理 電路包3有-處賴組，以及—齡單元，耦接至該處理模組。 於依序魏自-視絲源之·個魏酶+，該處理模組比較 —基準®框與其它視訊®框的至少-部分以產生-鱗，該遮罩 :、有了此對應於至少—標之至少—候選區域。該處理模組可對 該遮罩進行後處理以修正該至少一候選區域，並可將該至少一候 選區域中之至少一者選擇為標誌區域以供指出該至少一標諸。另 • 外，該儲存單元係用來儲存一指標陣列（indicatorarray)，該指標 陣列具有分珊應於每-視訊_巾不關位置之複數個指標 (indicator)其中母一指標指出目前於一位置是否被判定可能具 有一標誌。 a 【實施方式】 請參考第1圖，第1圖為依據本發明一實施例之一標誌（1〇g〇) 處理方法100的流程圖。標諸處理方法1⑻可藉由使用軟體、硬 體、及/或初體架構而被實施於一視訊（video)播放裝置（未顯 不）之至少一部分。例如：該視訊播放裝置可為一數位多用途光 碟片（Digital VersatileDisc，DVD)播放器，執行標誌處理方法1〇〇 之一標誌、處理電路（未顯示）係置於該視訊播放裝置中且被輕接 至該視訊播放裝置之MPEG解碼器，其中該標誌處理電路係用來 處理原本就被混合於一節目中之至少一標誌，而該節目係儲存於 该數位多用途光碟片播放器所存取（access)之一數位多用途光碟 片。又例如·该視訊播放裝置可為一個人錄影機（pers〇naivideo 12890241289024 IX. Description of the invention: [Technical field to which the invention pertains] - The present invention relates to a video playback device, and more particularly to a logo processing method and circuit. [Prior Art] A logo is widely used in television programs to announce the copyright of the provider or producer of the television program. A disc, such as a digital versatile disc (Digital Versatile D-DVD), may also have a similar use in one of the programs recorded in the program. Since the logo in the program is not directly responsive to the program β, and the logo sometimes partially blocks the particular video object displayed in the program (vide〇Gb(10), most viewers watching the program are occupying the television or the display. One of the glory/panel #秘_ domain trouble. Correction, if the county occupies the same corner in most of the coffee, especially for those who use the electro-polymer display panel 34 fiber (10) (four) one, the electric job or display The wire on the plasma display panel shows that the factory is not easy to burn or the performance is better than other display units. One of the first targets in the rsr TV program may be removed. This is because The first standard and the content of the program are transmitted from the service R to the wireless or tree transmission, and the silk is mixed in a 1289024. However, if the digital TV program has - the second standard is mixed in the program The identification removal architecture according to the above specifications is always invalid for the second label. SUMMARY OF THE INVENTION The object of the present invention is to provide a flag, (lGgG) processing method and circuit for detecting A further object of the present invention is to provide at least one mark of the program content. Another object of the present invention is to provide a mark processing method and circuit for changing at least one color that is originally mixed with at least one mark of the program content. The method of exposing a standard domain method includes comparing a reference frame (baseftame) with other video frames in a plurality of video frames (vide0frame) sequentially received from a video source (videosource). At least a portion of the frame to produce a mask having at least one candidate region that may correspond to at least one of the markers. The flag processing method further includes post processing the mask (P〇stpr〇) CeSSing) to modify the at least one candidate region, and select at least one of the at least one candidate region as a flag region for indicating the at least one flag. According to an embodiment of the invention, a flag processing method is disclosed The flag processing method includes detecting at least one flag, and changing at least one color of the at least one flag 1289024. According to the present invention - an embodiment The dew-type lining processing circuit, the standard processing circuit package 3 has a reliance group, and an ageing unit, coupled to the processing module. In accordance with the order Wei self-vision source, a Wei enzyme +, the The processing module compares the reference frame with at least a portion of the other video frame to produce a scale, the mask: having the corresponding at least the target at least the candidate region. The processing module can mask the mask Performing post-processing to modify the at least one candidate region, and selecting at least one of the at least one candidate region as the flag region for indicating the at least one target. In addition, the storage unit is configured to store an indicator Array (indicatorarray), the indicator array has a plurality of indicators that should be placed in the position of each video-indicator, wherein the parent-indicator indicates whether a flag is currently determined to have a flag at a location. [Embodiment] Please refer to Fig. 1. Fig. 1 is a flow chart showing a method 100 for processing a flag (1〇g〇) according to an embodiment of the present invention. The method 1(8) can be implemented on at least a portion of a video playback device (not shown) by using a software, hardware, and/or a primary architecture. For example, the video playback device can be a digital Versatile Disc (DVD) player, and one of the flag processing methods is executed. A processing circuit (not shown) is placed in the video playback device and is Lightly connected to the MPEG decoder of the video playback device, wherein the flag processing circuit is configured to process at least one flag that is originally mixed in a program, and the program is stored in the digital multi-purpose optical disc player Access one of the digital multi-purpose discs. For example, the video playback device can be a personal video recorder (pers〇naivideo 1289024
Recorder，PVR)，執行標誌處理方法1〇〇之一標誌處理電路（未顯 不）係置於該視訊播放裝置中且被耦接至該視訊播放裝置之記錄 緩衝器（recordingbuffer)，其中該標誌處理電路係用來處理原本 就被混合於一節目中之至少一標誌，而該節目係接收自該個人錄 影機之一電視調諧器。 “為處理方法1〇〇包含有偵測至少一標諸（例如：步驟U〇至 步驟140)，以及改變該至少一標誌（即該/該些標誌）之至少一 色彩（例如：步驟150)。在此，改變該至少一標誌之該至少一色 彩（即該/該些色彩）之典型實施方式，係週期性地或隨機地改 、邊ΰ亥至少一標誌之至少一部分之色彩及/或亮度。於是，就可避 免燒毁用來顯不該標誌、之顯示單元。 於步驟110中，於依序接收自一視訊來源（例如：一數位多用 途光碟片或一電視廣播系統）之複數個視訊圖框（videc)frame) FI、F2、F3、…^!!、···等當中，比較一基準圖框（baseframe) F1與其它視訊圖框的至少一部分以產生一遮罩（mask)，該遮罩 具有可能對應於至少一標誌之至少一候選區域。針對每一像素來 比較該基準圖框與該些其它視訊圖框之每一者，可應用於本發 明，其中要達到該遮罩之一穩定狀態可能要花許多時間。另外， 於一時期内自複數個視訊圖框選擇一視訊圖框（例如：針對每N 個視訊圖框，選擇一視訊圖框）以形成該些其它視訊圖框之該至 ’ 4刀，亦可應用於本發明。依據本實施例，對該些視訊圖框 1289024 進行鏡頭偵測（Shotdetection)以辨識該些視訊圖框之複數個鏡頭 (shot)，以及自該些視訊圖框中選擇對應於每一鏡頭之一關鍵圖 框（key frame)來代表該鏡頭，均可應用於步驟11〇。於是，該些 其它視訊圖框之該至少一部分包含有該些鏡頭之關鍵圖框。 請參考第2圖，第2圖為依據本發明一實施例之指標陣列 (indicatorarray)的變化之示意圖，其中該指標陣列可被應用於 第1圖所示之實施例。在此，該指標陣列可被用來代表第丨圖所 示之步驟110中所提到之遮罩。具有兩狀態，例如 :“0” 與 “1” 之 雙態（bi_state)指標（即該指標陣列或矩陣之元素）可被應用於 本發明，以藉由使用該兩狀態中之一者來代表該遮罩中之該至少 一候選區域（即該/該些候選區域）。依據本實施例，具有多狀態， 例如：“〇，，、“Γ，、“2”、“3，，、與 “4”之多態（multiple-state)指標 係被用來代表該遮罩中之該至少一候選區域，其中一指標之大小 代表一對應位置上具有一標誌之可能性。於更新該遮罩時，該指 標陣列中各指標之狀態可隨著改變。 將該指標陣列應用於第1圖所示之實施例，則步驟11〇另包含 有分別比較基準圖框F1之複數個區塊（bloek)與上述之該些其它 視訊圖框的該至少一部分之一特定視訊圖框之對應區塊，其中每 一區塊包含有包含該區塊之圖框之至少一像素；例如：該區塊具 有單一像素’即該區塊係為1乘以1大小之區塊；又例如：該區 塊具有複數個像素，則該區塊有可能是2乘以2大小之區塊、4 1289024 乘以4大小之區塊、8乘以8大小之區塊…等。此外，於步驟⑽ 中，依據對應於該特定視訊圖框之該些區塊之最新的比較結果來 更新該遮罩尤導致该遮罩隨時間來變化，例如：第3固所一之 變化。 如第3圖所示’一偵測區域係被用來排除於一第1迴圈中該此 視訊圖框之間對邊緣（margin)之比較，並進一步排除該此視訊圖 框之間多餘的比較。該"ί貞測區域之初始大小可被決定為小於一視 訊圖框之大小，以排除該視訊圖框之邊緣。另外，該初始大小另 可依據至少一視訊圖框（例如：基準圖框F1)上之一黑暗邊緣债 測（darkmargindetection)來決定或改變，以排除黑暗邊緣；黑暗 邊緣通常是節目影像與視訊圖框之間不同的晝面比例（aspect ratio)所造成，例如：若具有16比9的晝面比例之節目影像被記 錄於具有4比3的晝面比例之視訊圖框，則該些節目影像之上方 與下方就典型地出現黑暗邊緣。 隨著時間流逝，一旦一區塊由於該區塊對時間之變化而被判定 沒有標誌於其内或其上，該區塊就被標示為一無標誌區塊，這表 示在第3圖所示接下來的迴圈中，關於該區塊之進一步比較是不 需要的。另外，該些視訊圖框之間的比較之典型實施方式是比較 該些視訊圖框之像素值；不同的實施選擇可應用於本實施例。 依據一實施選擇，每一區塊包含有單一像素，即每一區塊係為 1289024 1乘以1大小之區塊，而步驟11〇則比較基準圖框^之每一區塊 之該像素之一像素值與該特定視訊圖框之一對應區塊之該像素之 一像素值，以產生該些最新的比較結果中之一者。依據上述之實 施選擇之一種變化，步驟U0包含有計算該基準圖框之每一區塊 之邊像素之該像素值與該特定視訊圖框之該對應區塊之該像素之 邊像素值之間之一差值，以及比較該差值與一門榣值 。在此，使 用該門檻值是為了避免雜訊所導致錯誤的影響。 依據另一實施選擇，每一區塊包含有複數個像素，例如：每一 區塊係為2乘以2大小之區塊，而步驟11〇則比較該基準圖框之 每一區塊之像素的像素值之一平均值與該特定視訊圖框之一對應 區塊之像素的像素值之-平均值，以產生該些最新的比較結果中 之一者。依據本實施選擇之一種變化，步驟11〇包含有計算該基 準圖框之每一區塊之像素的像素值之該平均值與該特定視訊圖框 之該對應區塊之像素的像素值之該平均值之間之一差值，以及比 較該差值與一門檻值。 依據另一實施選擇，每一區塊包含有複數個像素，例如：每一 區塊係為2乘以2大小之區塊，而步驟11〇則分別比較該基準圖 框之每—區塊之像素的至少一部紅像素值與該特定視訊圖框之 一對應區塊之像素的至少一部分之像素值，以計算該對應區塊 中、相對於該基準圖框之該區塊未被改變的像素之數量。步驟11〇 另依據未被改變的像素之該數量以及每一區塊中像素之數量，來 11 1289024 產生該些最新的比較結果中之一者。依據本實施選擇之一種變 化，步驟110分別計算該基準圖框之每一區塊之像素的該至少一 部分之像素值與該特定視訊圖框之該對應區塊之對應的像素之像 素值之間的差值，以及分別比較該些差值與一門檻值。 第4圖為依據本發明一實施例、結束比較基準圖框F1與該些 其它視訊圖框的該至少一部分之時間的示意圖，其中係定義了對 應於視訊圖框FI、F2、…、與F1800的圖框編號之一檢查區域 (inspectionregion)以及對應於視訊圖框 F1801、F1802、···、與 F3600的圖框編號之一可能結束區域，所以於步驟no中，基準圖 框F1與該些其它視訊圖框的該至少一部分之間的比較會於該檢查 區域内持續進行，並於該可能結束區域内結束。在該可能結束區 域内，步驟110與步驟120係被交替地執行，以藉由使用一持續 穩定計數器（continuously stable counter，Corn，stable counter )來檢 查該遮罩是否穩定，其中該持續穩定計數器之計數值係被監控 (monitor)以產生第4圖所示之曲線。依據本實施例，該持續穩 定計數器之計數值具有一初始值（例如：一百若於該特定視訊 圖框與基準圖框F1之間的比較之後，該遮罩未被改變或是不需要 改變，則该计數值就以一減少量（例如：一）來減少；否則，該 计數值就以大於該減少量之一增加量來增加。應用最簡單的實施 選擇於本實施例，則將該計數值以大於該減少量之一增加量來增 加可被代換成將該計數值重設（reset)為該初始值。於是，若當 圖框編號變⑽，該遮罩未被改變献不需要改變，_計數值 12 !289〇24 就減少並且最終會變成零。 於第4圖所示之實施例中，若該計數值於該可能結束區域内變 成零，則基準圖框F1與該些其它視訊圖框的該至少一部分之間的 比較就會結束。然而，若該計數值於該檢查區域内變成零，則只 要圖框編號還沒達到該檢查區域之末端（即該檢查區域與該可能 、、、"束區域之共同邊界），基準圖框F1與該些其它視訊圖框的該至 部分之間的比較憾會持續著。在某些情訂，若圖框編號 已達到該可能結束區域之末端（於本實施例係為該可能結束區域 之右侧邊界）碰計數值尚未變麟，則基準圖框F1與該些其它 视訊圖框的該至少—部分之間的比較會被強迫結束。 依據第4圖所不之實施例之一變化例，步驟12〇可被忽略，且 ’、要圖框編麵沒達到該檢查區域之末端，步驟iig係被重複地 執行；當圖框編號達到該檢查區域之末端時，步驟n〇中之比較 會被強迫結束。 依據第1圖所示之實施例，步驟130於典型實施方式中包含有 對每-候舰物卜獻（編iQn)運算贿正_選區域之 輪靡，以及把複數個相鄰候選區域聚集（gr〇uping)為一候選區域 以修正該些舰域之數量。改變該至少—標財至少—者之至少 -色衫。對該遮罩進行該後處理，是為了修正該至少—標誌以解 、一、㈣由雜崎引起之問題。於本實施例中，每一候選區 13 1289024 域係以-特定索引（例如：一特定數字）來標示，以供於後續步 驟中進-步儀。依據上述實施例之—變化例，若存在約為兩個 或三個像2的大小之-㈣區域，_候親域被視為過小，故 不以-特定索引來標補候縣域；此賴親域係典型地因雜 訊而產生。 於步驟130中，將該至少一候選區域中之至少-者選擇為標諸 區域以供指出（indicate)該至少一標誌、。在典型的實施方式中， 步驟14G涉及—位置分析程細及—大小分析程序，其中進行該 位置分析程序與該大小分析程序之順序可改變。以第5圖為例了 候選區域502、5〇4、與505於該位置分析程序中被剔除，而不被 標示為標麵域；這是因為錄置分析程料，任健選區域只 要被第5圖所示之剔除、線5U、512、513、切、切、切中之任 -者穿越’就應予剔除。另一方面，於該大小分析程序中對於 保留下來的候選區域’例如：第5圖所示之候選區域則、汹、 與506 ’則按照它們的大小來排序。於是，前_最大候選區域 係被選擇為標麵域，其巾m為—正整數。例如：若@等於二 則候選區域训與5〇3就被獅為標誌、區域。因此，分別由候 區域501與503所代表之標誌就被偵測到了。 /於步驟150中，步驟140所_到之該至少一標誌之至少一色 彩可被改變。如前面所述，去除或隱藏（eGneeaiing)該至少_標 誌可於步驟150中進行。 不 14 1289024 於步驟160 t，於該遮罩之每一標諸區域中，被镇測到之標誌 均被檢查。若於步驟170中依據步驟160之一計算結果偵測到一 新的標誌，則重新進入步驟11〇 ;否則，重新進入步驟16〇。 以上所述僅為本發明之較佳實施例，凡依本發明申請專利範圍 所做之均等變化與修飾，皆應屬本發明之涵蓋範圍。 【圖式簡單說明】 • 第1圖為依據本發明一實施例之一標誌(logo)處理方法的流程圖。 第2圖為依據本發明一實施例之指標陣列（indicat〇rafray)的變 化之示意圖，其中該指標陣列可被用來代表第1圖所示之步驟 110中所提到之遮罩（mask)。 第3圖為依據本發明一實施例之一遮罩對時間的變化之示意圖。 第4圖為依據本發明一實施例、結束比較基準圖框（baseframe) 與其它視訊圖框的至少一部分之時間的示意圖。 φ 第5圖為依據本發明一實施例、涉及第1圖所示之步驟14〇之一 位置分析程序。 15 1289024 【主要元件符號說明】 100標諸處理方法 100S，110, 120, 130, 140, 150, 160, 170 步驟 F1，F2，F3，F4,…，Fn視訊圖框 501，502, 503, 504, 505, 506 候選區域 511，512, 513, 521，522, 523 剔除線Recorder (PVR), a flag processing method (not shown) is disposed in the video playback device and coupled to a recording buffer of the video playback device, wherein the flag The processing circuitry is operative to process at least one of the indicia that would otherwise be mixed into a program received from a television tuner of the personal video recorder. "To process method 1 〇〇 includes detecting at least one target (eg, step U to step 140), and changing at least one color of the at least one flag (ie, the/the flags) (eg, step 150) Here, a typical embodiment of changing the at least one color (ie, the color) of the at least one mark periodically or randomly changes the color of at least a portion of at least one of the marks and/or Brightness. Therefore, it is possible to avoid burning the display unit for displaying the logo. In step 110, the plurality of video sources (for example, a digital multi-purpose optical disc or a television broadcasting system) are sequentially received. Video frame (videc) frame FI, F2, F3, ...^!!, ..., etc., compare a base frame F1 with at least a portion of other video frames to create a mask (mask The mask has at least one candidate region that may correspond to at least one of the markers. Comparing the reference frame and each of the other video frames for each pixel may be applied to the present invention, wherein the mask is One of the masks is stable It takes a lot of time. In addition, a video frame is selected from a plurality of video frames in a period (for example, for each N video frames, a video frame is selected) to form the other video frames. The 4 knives can also be applied to the present invention. According to the embodiment, the video detection frame 1289024 performs a shot detection (Shotdetection) to identify a plurality of shots of the video frames, and the video from the video frames. The selection of a key frame corresponding to each lens in the frame to represent the lens can be applied to step 11 〇. Therefore, at least a part of the other video frames contain the key of the lens. Please refer to FIG. 2, which is a schematic diagram showing changes of an indicator array according to an embodiment of the present invention, wherein the indicator array can be applied to the embodiment shown in FIG. 1. The indicator array can be used to represent the mask mentioned in step 110 of the figure. There are two states, for example: a bistate (bi_state) indicator of "0" and "1" (ie, the indicator array or Element of matrix Can be applied to the present invention to represent the at least one candidate region (i.e., the/the candidate regions) in the mask by using one of the two states. According to the embodiment, having multiple states, For example, the "multiple-state" indicator of "〇,,,"Γ,, "2", "3,", and "4" is used to represent the at least one candidate region in the mask, wherein The size of an indicator represents the likelihood of having a flag at a corresponding location. When the mask is updated, the state of each indicator in the array of indicators can change. Applying the indicator array to the embodiment shown in FIG. 1 , step 11 further includes comparing a plurality of bloeks of the comparison reference frame F1 with the at least a part of the other video frames. a corresponding block of a specific video frame, wherein each block includes at least one pixel including a frame of the block; for example, the block has a single pixel 'that is, the block is 1 by 1 size Block; for example: the block has a plurality of pixels, then the block may be a block of 2 times 2 size, 4 1289024 times 4 blocks, 8 times 8 blocks...etc. . In addition, in step (10), updating the mask according to the latest comparison result of the blocks corresponding to the specific video frame particularly causes the mask to change with time, for example, a change of the third solid. As shown in Figure 3, a detection area is used to exclude the margin between the video frames in a first loop, and further eliminates the extra between the video frames. Comparison. The initial size of the "measurement area can be determined to be smaller than the size of a video frame to exclude the edge of the video frame. In addition, the initial size may be determined or changed according to one of the dark margin detections of at least one video frame (eg, reference frame F1) to exclude dark edges; the dark edges are usually program images and video images. The difference in the aspect ratio between the frames, for example, if a program image having a face ratio of 16 to 9 is recorded in a video frame having a face ratio of 4 to 3, the program images are Dark edges typically appear above and below. As time passes, once a block is determined to be unmarked within or on the block due to changes in time, the block is marked as an unmarked block, as indicated in Figure 3. In the next loop, a further comparison of the block is not needed. In addition, a typical implementation of the comparison between the video frames is to compare the pixel values of the video frames; different implementation options are applicable to the present embodiment. According to an implementation choice, each block includes a single pixel, that is, each block is 1289024 1 times 1 size block, and step 11〇 compares the pixel of each block of the reference frame ^ One pixel value corresponds to one of the pixels of the block corresponding to one of the specific video frames to generate one of the latest comparison results. According to a variation of the implementation choices described above, step U0 includes calculating a pixel value of a pixel of a side of each block of the reference frame and a pixel value of a pixel of the pixel of the corresponding block of the specific video frame. One of the differences, and the difference is compared to a threshold. Here, the threshold is used to avoid the effects of errors caused by noise. According to another implementation, each block includes a plurality of pixels, for example, each block is a block of 2 by 2 size, and step 11〇 compares pixels of each block of the reference frame. One of the pixel values averages and the average of the pixel values of the pixels of the corresponding block of the particular video frame to produce one of the most recent comparison results. According to a variation of the selection of the implementation, the step 11〇 includes calculating the average value of the pixel values of the pixels of each block of the reference frame and the pixel values of the pixels of the corresponding block of the specific video frame. The difference between the average values, and the comparison of the difference with a threshold. According to another implementation, each block includes a plurality of pixels, for example, each block is a block of 2 times 2 size, and step 11〇 compares each block of the reference frame separately. Converting at least one red pixel value of the pixel to a pixel value of at least a portion of a pixel of the block corresponding to one of the specific video frames to calculate that the block in the corresponding block relative to the reference frame is unchanged. The number of pixels. Step 11 〇 According to the number of pixels that have not been changed and the number of pixels in each block, 11 1289024 generates one of the latest comparison results. According to a variation of the selection of the implementation, step 110 respectively calculates a pixel value of the at least one portion of the pixel of each block of the reference frame and a pixel value of a corresponding pixel of the corresponding block of the specific video frame. The difference, and the difference between the difference and the threshold. 4 is a schematic diagram of ending the comparison of the reference frame F1 and the at least part of the other video frames according to an embodiment of the present invention, wherein the video frames FI, F2, ..., and F1800 are defined. One of the frame number inspection areas (inspectionregion) and one of the frame numbers corresponding to the video frame F1801, F1802, ..., and F3600 may end the area, so in step no, the reference frame F1 and the A comparison between the at least a portion of the other video frames will continue within the inspection region and end within the possible end region. In the possible end region, steps 110 and 120 are alternately performed to check whether the mask is stable by using a continuously stable counter (Corn, stable counter), wherein the persistent stable counter is The count value is monitored to produce the curve shown in FIG. According to this embodiment, the count value of the persistent stable counter has an initial value (for example, one hundred if the comparison between the specific video frame and the reference frame F1 is not changed or does not need to be changed. The count value is decreased by a decrease amount (for example, one); otherwise, the count value is increased by an increase amount larger than the decrease amount. The simplest implementation is selected in the embodiment, and the The count value is increased by an increase greater than the decrease amount, and may be replaced by resetting the count value to the initial value. Thus, if the frame number is changed (10), the mask is not changed. Need to change, the _count value 12 ! 289 〇 24 is reduced and will eventually become zero. In the embodiment shown in Figure 4, if the count value becomes zero in the possible end region, the reference frame F1 and the The comparison between the at least one portion of the other video frames ends. However, if the count value becomes zero in the inspection area, then the frame number has not yet reached the end of the inspection area (ie, the inspection area is The possibility, , "the common boundary of the bundle region), the comparison between the reference frame F1 and the other portions of the other video frames will continue. In some cases, if the frame number has reached the possible end region At the end (in this embodiment, the right edge of the possible end region), the count value has not changed, and the comparison between the reference frame F1 and the at least part of the other video frames is forced. According to a variation of the embodiment of FIG. 4, step 12〇 can be ignored, and “the frame immette does not reach the end of the inspection area, and the step iig is repeatedly executed; when the frame is When the number reaches the end of the inspection area, the comparison in step n〇 is forced to end. According to the embodiment shown in Fig. 1, step 130 includes in the exemplary embodiment a pair of items (iQn) Calculating the rim of the selected area, and arranging a plurality of adjacent candidate areas as a candidate area to correct the number of the ships. Changing at least - at least - at least - Color shirt. The back of the mask In order to correct the problem that at least the flag is caused by the solution, one, and (four) by the miscellaneous. In this embodiment, each candidate zone 13 1289024 domain is marked with a specific index (for example, a specific number) to For the step-by-step procedure in the subsequent steps. According to the variation of the above embodiment, if there is a - (four) region of the size of about two or three images of 2, the _ candidate domain is considered too small, so it is not - A specific index is used to qualify the county domain; the neighboring domain is typically generated by noise. In step 130, at least one of the at least one candidate region is selected as the target region for indicating the at least In a typical embodiment, step 14G involves a location analysis process and a size analysis program, wherein the order in which the location analysis program and the size analysis program are performed can be changed. Taking the fifth picture as an example, the candidate areas 502, 5〇4, and 505 are eliminated in the location analysis program, and are not marked as the label field; this is because the recording analysis material, any healthy selection area is only The culling, line 5U, 512, 513, cutting, cutting, and cutting shown in Figure 5 should be eliminated. On the other hand, in the size analysis program, for the reserved candidate regions ', for example, the candidate regions shown in Fig. 5, 汹, and 506 ' are sorted according to their sizes. Thus, the pre-maximum candidate region is selected as the facet field, and the towel m is a positive integer. For example, if @ is equal to two, the candidate area training and 5〇3 are marked by the lion and the area. Therefore, the flags represented by the waiting areas 501 and 503, respectively, are detected. / In step 150, at least one of the at least one flag to which step 140 is to be changed may be changed. As previously described, the at least _ flag may be removed or hidden (eGneeaiing) in step 150. No. 14 1289024 In step 160 t, the marked flag is inspected in each of the marked areas of the mask. If a new flag is detected in accordance with the result of the calculation in step 170 in step 170, then step 11 is re-entered; otherwise, step 16 is re-entered. The above are only the preferred embodiments of the present invention, and all changes and modifications made to the scope of the present invention should fall within the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a flow chart showing a method of processing a logo according to an embodiment of the present invention. 2 is a schematic diagram showing changes in an index array (indicat〇rafray) according to an embodiment of the present invention, wherein the index array can be used to represent the mask mentioned in step 110 shown in FIG. . Figure 3 is a schematic illustration of a change in mask over time in accordance with one embodiment of the present invention. Figure 4 is a diagram showing the timing of ending at least a portion of a base frame and other video frames in accordance with an embodiment of the present invention. φ Fig. 5 is a view showing a position analysis procedure relating to step 14 of Fig. 1 according to an embodiment of the present invention. 15 1289024 [Explanation of main component symbols] 100 standard processing methods 100S, 110, 120, 130, 140, 150, 160, 170 Steps F1, F2, F3, F4, ..., Fn video frames 501, 502, 503, 504 , 505, 506 candidate areas 511, 512, 513, 521, 522, 523 culling lines
1616