正替換頁 1246861 九、發明說明： 【發明所屬之技術領域】 本發明係有關於視訊壓縮技術，更詳而言之，係有關 於一種用以有效率的壓縮輸入視訊圖框之視訊編碼與解碼 方法及其裝置。 【先前技術】 廣泛應用在視訊監視系統的視訊壓縮技術係依據國際 視訊壓縮標準，亦即H· 263、動晝專家群（motion picture experts group ; MPEG)等等。採用依據 Η· 263 或 MpEG 吟 壓縮標準之傳統視訊編碼器係顯示於第丨圖中。該編碼器 透過執行框内編碼壓縮與框間編碼壓縮將原始視訊圖框予 以編碼。 第2圖顯示藉由第丨圖所示的視訊編碼器所生成的圖 框之排列。如前所述’習知的視訊編碼㈣依據框内壓縮 技術執行框内編碼並依據框間壓縮技術執行框間編 該框内編碼作業透過分離輸人視訊圖框至複數個區塊Positive replacement page 1246861 IX. Description of the invention: [Technical Field] The present invention relates to video compression technology, and more particularly to video coding and decoding for efficient compression of input video frames Method and apparatus therefor. [Prior Art] The video compression technology widely used in video surveillance systems is based on the international video compression standard, namely H.263, motion picture experts group (MPEG) and so on. A conventional video encoder based on the Η·263 or MpEG 压缩 compression standard is shown in the figure. The encoder encodes the original video frame by performing in-frame coding compression and inter-frame coding compression. Figure 2 shows the arrangement of the frames generated by the video encoder shown in the figure. As described above, the conventional video coding (4) performs intra-frame coding according to the in-frame compression technique and performs inter-frame coding according to the inter-frame compression technique. The intra-frame coding operation is performed by separating the input video frame to a plurality of blocks.
一致性的以圖框表示。 書中所用的術語圖場與圖框 當框間編碼作業係依據預測介於先前Consistency is represented by a box. The term field and frame used in the book when the inter-frame coding operation is based on predictions
的增加壓縮效率。依據所應用之該動態 的增加壓縮效率。 則圖框與依據來自 執行，因此可有效 預測方法分離該框 124686i 〆)正替換頁j 間編碼至测圖框（p @框）編碼與雙向圖框（B圖框）編 瑪。 =第2圖所示，該預測圖框表示圖框係依據由先前圖 框^向制予以編碼，而該雙向雜表示圖框係依據由 先前圖框以及次—圖框的前向預測予以編碼。依據視訊編 碼中的圖框類型之壓縮效率於丨圖框中相對較低，然於p 圖框以f B圖框中則相對較高。當然、，於B圖框中的壓縮 效率又高於P圖框中的壓縮效率。 …因為來自視訊編碼中複數個圖框的動態預測係依據介 於複數個圖框中相㈣多餘者，丨圖框典型的制於場景 發生轉換時。 該視汛監視系統根本上將監視視訊圖框予以編碼並儲 :或摘該編碼後之視訊圖框。該視訊監視祕需要用以 執行f記錄魏_前_放、後向播放以及搜尋等作業 的功能。該視訊監視系統要求記錄該視訊 ;錄的視訊圖框等功能。為滿u援前述特殊的^功 月匕，而將I圖框插入特定的間隔。 該I圖框間隔係與解碼系統相關，但該丨圖框係用於 大約10 11框間隔。為了使後向播放可於該解碼系統中執 行’將介於該I 所㈣圖框予崎碼並該解碼後的 圖框予以儲存。接著，該解碼系統必須在視訊解碼的順序 下以反序方式依序顯示該解碼後的圖框。當該丨圖框間隔 非常長時’所儲存的圖框數量會增加，故需要 時間。 考量到前向播放、後向播放以及搜尋作業，習知的視 1246861 於此产系統反覆的於數個圖框之預定間隔中利用i圖框。 生的产兄下’會有一個問題，亦即儘管在沒有場景變更發 、月/兄下該I圖框會被不必要插入，進而導致壓縮效率 的降低。 【發明内容】 址為解決上述習知技術之缺點，本發明之一目的在於提 1共回種可以減少重製與搜尋記錄的視訊圖框所需之圖框1 、囷忙數里之視汛編碼/解碼裝置及其方法，藉以增加記錄 的視訊圖樞之壓縮效率。 ’、 么本發明之另一目的在於提供一種可透過於視訊監視系 ^旦減J重製與搜尋記錄的視訊圖框所需之圖框I的圖框 1 =增加記錄的視訊圖框效率，且可增加壓縮效率並可 j日獨間喊介於記錄的視訊圖框間亮度變化藉由校正 避免視訊圖框品質降低之視訊糾/解碼錢及其方法。 依據本發明之_型態，前述以及其他的目的可藉由 4匡内編竭與郎編碼輸人她目生成 視訊編碼方法的提供扣實現，該方純含：讀机之 將，内編碼視訊圖框儲存為參考視訊圖框；以及 於每-框⑽碼咖點’參考該參考視訊 測的將科視訊圖框料編碼。 本發明可藉由考量配置有用以取得監視視訊的攝影機之 I視系統的特性並於每—框、 測以增加資料M縮率。 ㉟心向預 依據本發明之另—型態’前述以及其他的目的可藉由用於 1246861 框内編碼與插pq HU: 4、為碼輪入視訊圖框並生成編碼資料流之視 了:法的提供予以實現，該方法包含： 白=一，編石馬視訊圖框儲存為參考視訊圖框；以及 圖框的每瑪時間點開始，該框内編碼依據現時視訊 框區塊差！^塊以及相應之先前參考視訊圖框區塊的圖 今户^子在透過選擇的跨越予以執行並將用以指示 …交換位元插入至相應區塊的編碼資料間隔。 作業 明之該些型態，當必須執行該框内編碼 之具有視考視訊圖框以及峨^ 之區塊予以框2 且僅針對具有圖框區塊差異 視加框扩肉碼。因此’該創新的編碼方法與將全部 於此提出供：： 較可增加屢縮效率。 參考視訊圖框改變圖框區塊係與先前編碼 該參考視訊圖框區塊在解:期間變圖框區塊相同 框可被正常的解碼。〜s 1叟♦’據此-視訊圖 之又一型態，該視訊圖框編碼以 法’係匕3.虽框内編碼視訊圖框之每一 馬方 依據該交換位元值將相廊區 ‘凡成解碼， 碼，或存取相應該參考^予以解 訊解碼作業。於此提出供參考，為避』執行視 品質降低，較佳的當遇到敗要求 ς 2影像 框。為此，當設定時間達到時將該現時視見訊圖 編碼，並將該框内編焉視訊圖框儲存成該^予从框内 更新該參考視訊圖框。 Μ >考視乱圖框俾 1246861 _幸乂佳的，當介於該芩考視訊圖框與該現時視訊圖框間 焭度訊息變化總量超過預設的臨界值時，框内編碼該現時 視矾圖框並將該框内編碼視訊圖框儲存為參考視訊圖框俾 更新遠參考視訊圖框。 【實施方式】 以下將伴隨所附的圖式更具體的說明本發明之較佳實 施例。独τ的制巾，為職_本發明之躲所在結 合於本實施例之習知功能與結構將予以省略。 本發明可應用於用以編碼透過攝影機所輸入之視訊圖 框的視ifi監視线，儲存該編碼視訊圖框至記錄媒體，將 ，存於該e錄雜之編碼視訊圖框解碼，涵示該解碼視 2框此外，本發明可應用於用以編碼視訊圖框並透過 1線/無線親傳送該編碼視訊圖框之系統，或用以解碼透 k該有線/無線頻道輸入的編碼視訊圖框之系統等等。 第/®剌以顯錄據本發明之—實翻的視訊編碼 太發明衣置之方塊圖。更具體而言’第3圖係用以顯示依據 电月之視訊監視系統的視訊編媽/解碼裝置之方塊圖。 第3圖，視訊編碼器4G選擇性地將輸入視訊圖框予 框内、、扁石馬以及框間編碼並生成編碼資料流。 親見°Κ、為碼益40儲存框内編碼視訊圖框以作為參考 編碼作業執行完成，該視訊編刪 =考、考視_框並前向_地將現時視訊圖框編 碼時月二另:Γ·該視訊編碼器40於框内編 將現％視㈣框予以框内編碼。此外，於框間編 1246861 烏碼器40依據介於該現時視訊圖框每-二摞=i先前參考視訊圖框區塊間圖*匡區塊差異的存 性:跨越框間編碼作業。將用以指示該跨越存在的 間=凡插入至相應區塊的編碼資料間隔’據此可執行框 依據本發明之又一實施例，該視訊編碼器 框内編碼視訊圖框至參考圖框記憶單元7〇接 目 視4:介於該現時視訊圖框每-區 >考視汛圖框區塊間的圖框區塊差異 以L該:訊編碼器4°提供該現時視訊圖框心 元插人至相應區塊的柄資料間隔，據此可執 $圖出供參考，該視訊編碼1140 _介於該現時視 :心u母一Γ:與相應的參考視訊圖框區塊間之絕對誤差 = Abs°luteDifier⑽咖）是否超過預設來 因此，該視訊編碼n 40可依據_斷@ 、 訊圖框區塊間的圖框區塊差里 果仏查兩視 :視訊_度與該第一框_亮度; 見訊圖框極度的取決於時間間隔與環就變化。因 +為使該視Λ圖框的塵縮效率可連續並適 種用以克服該依據時間間隔或環境變;匕之極度亮度 當設定時間到達時，該視訊編碼器4 〇 現時視訊圖框並儲存該框内編碼視訊圖框^為财考: 10 I246861£21^ 存再-考_ 變化總量超過預設的：見化視訊圖框間之亮度訊息 以作為新參考視气圖框=亚儲存該框内編碼視訊圖框 若監親= 更新該參考視訊圖框。 、編碼該現時度===設的臨界值，框内 新參考=圖框藉以更新該=視訊圖框《作為 自該視訊出 之視所輪出 !_間係為第一框内編碼視訊圖框: %境變化予以更新。 依據4間間隔或 依據本發明之一實施例，視訊編 訊編碼器40以及該參考圖框記憶單元7()：；=該視 包括視訊解碼器6。。-視訊監視系統二 視訊'抓儲存單元5〇、亮度感測器等等所組成。。卩乂由 ，訊解碼器60係構成該視訊編瑪 :用：據該交換位元值將相應區塊之待解^ 予以解碼。可替代的，該視訊解碼器6〇可存取相菩2 碼區塊的茶考視訊圖框之解碼資料以回復原始視訊圖框。 以下，依據本發明之視訊編碼以及解碼方法將 施例予以說明。由於框内編碼與框間編碼係習知只 細節描述將予以省略。 故 11 !246861 第一實施例 由於該視訊監視系統之攝影機通常係固定的，透過該 固定攝影機娜本質上具有相同減圖框特㈣視訊圖 框。考量該視訊圖框特性，由於場景並未變化故該視訊監 視系統無須預先生成I圖框。除非該視訊監視系統要求！ 圖框以執行前向播放、後向播放以及搜尋等操作時。 因此本&明提出新的ρ!圖框類型用以於增加壓縮效 f時取代該I圖框。當框⑽錢行時生成該ρι圖框以取 代該I圖框。當壓縮該視訊圖框後，基於來自第一框内編 =視_框之前向預測將作為現時輸人視訊圖框而生成的 圖框所必須之「參考視訊圖框 應用於如第3圖所 二w考視關框係儲存並 管栗开日^ 圖框記憶單㈣等分離的圖框 二=先^的依據時間間隔與環境變化而再次更新。 編瑪視訊圖框，亦即該參考視訊圖框， 壓縮率。盘習知/馬π間”、、占可取得該存在的P圖框 可増加屋縮^喻貝料編碼方法相較本發明之第-實施例 透過該第一眘#Μ α 先前業解： 12 1246861 / , 6 框-動： 解碼。如前所述，透過視訊解圖框予以 打圖框解碼成原始圖框。"方法了將5亥别向預_碼 依據本發明之第一實施例， 時參考先前編碼參考視訊圖框向；碼作業 框，與習知的資料 預貝u馬現日寸視訊圖 加壓縮效率。於此提出2較創新的資料編碼方法可增 間間隔與環境變化而在該翏考視訊圖框依據時 下。〗兄义化而予以更新的情況將例示性的說明如 現時時該視訊編碼器仙可内部編碼 視訊=藉以更新“二==訊圖框以作為新參考 _現時====單：?°中的參考視訊 η士，拍βπ μ δ孔息變化總量超過預啤昨灵伯 时，框内編碼該現時 里μ預^界值 存成=訊圖框藉以更新 編碼視訊圖_ 瑪該化總量超過預設臨界值’框内編 新的參考视訊圖框，存該框内編碼視訊圖框以作為 新該參考视訊9 、祈該爹考視訊圖框。用以自動更 中。 回i、作業可應用於本發明之所有實施例 第二實施例 13 1246861! /.iF-fMwl 本考X明之第二實施例提供一種可 框區塊相關聯之交換位元以増加壓的= 馬視訊圖 框間編碼作業並生視法執行框内與 _為參考視訊圖框而儲存於“:訊圖框 行框間編碼時提供該現時= =間編:係依據介於該現時視訊圖 塊所存在的差里而選擇的區塊間圖框區 碼資：二曰不: 的交換位元插入至相應區塊的編 ，、、、枓間隔，據此完成該框間編碼。 :種用以判斷介於兩區塊間圖框區塊差異的差異之方 考判斷介於該現時視訊圖框的每一區塊與相應之泉 2關框區塊間的絕對誤差總和是否超過預設參考值予 執仃:若介於兩區塊間之圖框區塊差異不存在，將該交 显=70设置為「〇」。否則，若介於兩區塊間之圖框區塊差 /、存在，則將該交換位元設置為「1」。 視Λ解碼階段依據該交換位元值將相應圖框區塊之框 間編碼資料流予以解碼。可替代的，該視訊解碼階段存取 於該完成解碼參考視訊圖框中現時圖框區塊相關之解石馬資 料且接著執行正常視訊解碼作業。 σ §框間編碼作業執行完成時，可依據介於該現時圖框 區塊之區塊與相應之參考視訊圖框區塊間的圖框區塊存在 之差異選擇性地連同狀態訊息框間編碼現時圖框區塊。因 此’當針對不具有龍d塊差異之區塊僅有丨位元狀態訊 14Increase the compression efficiency. The compression efficiency is increased depending on the dynamics applied. Then the frame and the basis are from execution, so the effective prediction method separates the frame 124686i 〆) the replacement page j is encoded into the frame (p @ box) code and the bidirectional frame (B frame). = As shown in Fig. 2, the prediction frame indicates that the frame is encoded according to the previous frame, and the two-way hybrid frame is encoded according to the forward prediction of the previous frame and the secondary frame. . The compression efficiency according to the frame type in the video coding is relatively low in the frame, but the p frame is relatively high in the f B frame. Of course, the compression efficiency in the B frame is higher than the compression efficiency in the P frame. ...because the dynamic prediction from the multiple frames in the video coding is based on the multiple (4) extras in the multiple frames, the frame is typically used when the scene transitions. The video surveillance system essentially encodes and stores the surveillance video frame: or extracts the encoded video frame. This video surveillance secret needs to be used to perform functions such as recording _ pre-release, backward playback, and search. The video surveillance system requires recording of the video, recorded video frames and the like. Insert the I frame into a specific interval for the special purpose of the above. The I-frame interval is related to the decoding system, but the frame is used for approximately 10 11 frame intervals. In order for the backward playback to be performed in the decoding system, the frame that is in the I (4) frame and the decoded frame is stored. Then, the decoding system must sequentially display the decoded frame in reverse order in the order of video decoding. When the frame interval is very long, the number of frames stored will increase, so it takes time. Considering forward play, backward play, and search work, the conventional view 1246861 utilizes the i frame in the predetermined interval of several frames repeated by the production system. There will be a problem with the birth of the brother, that is, even if there is no scene change, the I frame will be inserted unnecessarily, resulting in a reduction in compression efficiency. SUMMARY OF THE INVENTION In order to solve the above-mentioned shortcomings of the prior art, one of the objects of the present invention is to provide a picture frame 1 that can reduce the number of video frames for re-creation and search recording, and the number of frames in the busy number. An encoding/decoding apparatus and method thereof for increasing the compression efficiency of a recorded video picture pivot. Another object of the present invention is to provide a frame 1 required for the video frame of the video surveillance system to reduce the J-reproduction and search recording, and to increase the efficiency of the recorded video frame. Moreover, the compression efficiency can be increased and the video brightness between the recorded video frames can be shouted by the correction of the video correction/decoding method and the method for avoiding the degradation of the video frame quality. According to the _ type of the present invention, the foregoing and other objects can be achieved by providing a deduction of the encoding method of the video encoding method in the 匡 , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , The frame is stored as a reference video frame; and the video frame is encoded in the reference frame for each frame (10). The present invention can be used to obtain the characteristics of the I-view system of the camera for monitoring video and to increase the data reduction rate in each frame. 35 Heart orientation according to another aspect of the present invention 'The foregoing and other objects can be achieved by using 126861 in-frame coding and inserting pq HU: 4. For the code wheel to enter the video frame and generate the encoded data stream: The method is provided, and the method comprises: white=1, the framed video frame is stored as a reference video frame; and the frame starts at a time point of each frame, and the code in the frame is based on the current video frame block difference! The block and the corresponding previous reference video block block are executed by the selected span and will be used to indicate that the swap bit is inserted into the coded data interval of the corresponding block. The types of operations are clearly defined. When the frame with the video frame and the frame code must be executed, the block with the frame and the frame must be framed 2 and only for the frame block difference. Therefore, the innovative coding method will be all proposed here:: It can increase the efficiency of the contraction. The reference video frame changes the frame block system and the previous code. The reference video frame block is decoded in the same frame during the solution: the variable frame block. ~s 1叟♦ 'According to this - another type of video map, the video frame encoding is based on the law 'system 3. Although each of the in-frame coded video frames will be in accordance with the exchange bit value The area 'failed to decode, code, or access the corresponding reference ^ to decode the decoding operation. It is hereby provided for reference, in order to avoid the implementation of the visual quality degradation, it is better to encounter the failure requirement 影像 2 image frame. To this end, the current view picture is encoded when the set time is reached, and the edited video frame in the frame is stored in the frame to update the reference picture frame. Μ >考乱乱框俾俾1246861 _ Fortunately, when the total amount of change in the information between the reference video frame and the current video frame exceeds a preset threshold, the frame encodes the The current view frame and the coded video frame in the frame are stored as a reference video frame and the remote reference video frame is updated. [Embodiment] Hereinafter, preferred embodiments of the present invention will be described more specifically with reference to the accompanying drawings. The unique function and structure of the present invention which is incorporated in the present embodiment will be omitted. The present invention is applicable to an audio ifi monitor line for encoding a video frame input through a camera, storing the coded video frame to a recording medium, and decoding the encoded video frame stored in the e-recording, indicating the The present invention can be applied to a system for encoding a video frame and transmitting the encoded video frame through a 1-line/wireless user, or a coded video frame for decoding the wired/wireless channel input. The system and so on. / / 剌 显 显 显 显 显 显 显 显 本 本 本 本 本 本 本 本 。 太 太 太 太 太 太 太More specifically, Fig. 3 is a block diagram showing a video editing/decoding device according to the video surveillance system of the electric moon. In Fig. 3, the video encoder 4G selectively encodes the input video frame into the frame, the slab and the frame, and generates an encoded data stream. The video frame is encoded as a reference coded work in the code box of the code 40, and the video code is deleted, the test is checked, and the current video frame is encoded in the forward direction. : The video encoder 40 encodes the current % (four) frame in the frame. In addition, the inter-frame code 1246861 U-coder 40 is based on the inter-frame coding operation of the current video frame per-two 摞 = i previous reference video frame block diagram * 匡 block difference. The coded data interval to be used to indicate the existence of the span = where the coded data interval is inserted into the corresponding block. According to another embodiment of the present invention, the video encoder frame encodes the video frame to the reference frame memory. The unit 7 is connected to the visual 4: the frame difference between the block of the current video frame and the frame of the frame is L: the encoder 4° provides the current video frame core Insert the data interval into the corresponding block, according to which the figure can be used for reference. The video code 1140 _ is between the current view: the heart and the mother: the absolute between the block and the corresponding reference video frame block. The error = Abs°luteDifier(10) coffee) is more than preset. Therefore, the video code n 40 can be checked according to the frame difference between the block and the frame of the message frame: video_degree and the first A frame _ brightness; the visibility of the frame is extremely dependent on the time interval and the ring changes. The video encoder 4 〇 视 视 因 因 因Save the coded video frame in the box ^For the financial test: 10 I246861£21^ Save again - test _ The total amount of change exceeds the preset: The brightness information between the video frames is used as the new reference gaze frame = Asia Store the coded video frame in the box. If the parent is = update the reference video frame. , code the current value === set threshold, the new reference in the frame = frame to update the = video frame "as the view from the video out of the round! _ is the first frame of the coded video map Box: % environment changes are updated. According to an embodiment of the present invention, the video encoding encoder 40 and the reference frame memory unit 7(): == the video decoder 6 is included. . - Video Surveillance System 2 Video 'catch storage unit 5 亮度, brightness sensor, etc. . The video decoder 60 constitutes the video semaphore: by: decoding the corresponding block to be decoded according to the exchange bit value. Alternatively, the video decoder 6 can access the decoded data of the tea test video frame of the phase 2 block to restore the original video frame. Hereinafter, the embodiment will be described in accordance with the video encoding and decoding method of the present invention. Since the in-frame coding and the inter-frame coding system are known, only the detailed description will be omitted. Therefore, 11 !246861 First Embodiment Since the camera of the video surveillance system is usually fixed, the fixed camera has essentially the same frame (four) video frame. Considering the characteristics of the video frame, the video surveillance system does not need to generate an I frame in advance because the scene has not changed. Unless the video surveillance system requires it! Frame to perform forward playback, backward playback, and search operations. Therefore, this & explicitly proposes a new ρ! frame type to replace the I frame when increasing the compression effect f. The ρι frame is generated when the box (10) is lined up to replace the I frame. After compressing the video frame, the "reference video frame" necessary for the frame generated from the first frame to be predicted to be the current input video frame is applied to the image frame as shown in FIG. The second w examines the frame and stores it and controls the chestnut opening. ^ The frame memory card (4) and other separated frames 2 = first ^ is updated again according to the time interval and environmental changes. The video frame, that is, the reference video Frame, compression ratio. Pan Xizhi/Ma π", the P frame that can obtain the existence can be added to the housing. The coding method is compared with the first embodiment of the present invention. Μ α Previous solution: 12 1246861 / , 6 Box-action: Decoding. As mentioned above, the frame is decoded into the original frame through the video solution frame. According to the first embodiment of the present invention, the method refers to the previous coded reference video frame direction; the code operation frame, and the conventional data pre-beautiful video Compression efficiency. It is proposed here that 2 more innovative data encoding methods can increase the interval and environmental changes in the reference video frame. The situation in which the brothers are renewed will be exemplified. If the video encoder is now able to internally encode the video = by updating the "two == frame" as a new reference_current ==== single: ?° In the reference video η, the total amount of βπ μ δ hole change exceeds the pre-Beer yesterday, the current code in the frame is stored in the current frame = the frame is updated to encode the video image _ 玛化The total number of reference video frames in the box exceeds the preset threshold value, and the video frame is encoded in the frame as the new reference video frame 9, which is used to automatically check the video frame. Back to i, the operation can be applied to all embodiments of the present invention. The second embodiment 13 1246861! /.iF-fMwl The second embodiment of the present invention provides a switchable block associated with the blockable block. The video coding operation between the video and the video is performed in the frame and the _ is the reference video frame and is stored in the ": frame between the frame of the message frame to provide the current = = inter-code: based on the current video map The block between the blocks selected in the difference between the blocks: the second bit: the exchange bit Enter the interval of the edit, , , and 相应 of the corresponding block, and complete the inter-frame coding accordingly. : The method used to determine the difference between the difference between the two blocks is determined whether the sum of absolute errors between each block of the current video frame and the corresponding spring 2 block is exceeded. Preset reference value: If the difference between the frame blocks between the two blocks does not exist, set the intersection = 70 to "〇". Otherwise, if the frame block between the two blocks is /, it is set to "1". The video decoding stage decodes the inter-frame encoded data stream of the corresponding frame block according to the exchange bit value. Alternatively, the video decoding stage accesses the calculus data associated with the current frame block in the decoded reference video frame and then performs a normal video decoding operation. When the execution of the inter-frame coding operation is completed, the inter-frame coding may be selectively performed according to the difference between the block between the block of the current frame block and the corresponding reference video frame block. The current frame block. Therefore, when there is only a bit status message for a block that does not have a difference in the d block.
息被編碼時，相較該現存框間編碼圖框該創新框間編碼圖 框可具有較高的壓縮效率。 第三實施例 本發明之第三實施例提供一種應用前述Pi圖框與交 換位元用以增加壓縮效率之方法。本發明之第三實施例將 伴隨所附之圖式詳細說明如下。 第4圖係為顯示依據本發明之一實施例之視訊編碼流 程之流程圖；第5圖係為顯示依據本發明之一實施例用以 將視圖框區塊解碼之流程的流程圖；而第6與7圖則係 為顯示透過依據本發明之視訊編碼器40所生成的圖框排 列例示。 請參閱第4圖，該視訊編碼器40於該視訊編碼期間將 輪入視訊圖框予以第一框内編碼（於步驟200中）。當該視 汛圖框如前述般予以編碼時，該第_框内編碼視訊圖框儲 存於該參考圖框記憶單元70中以作為參考視訊圖框（於牛 驟 210 中）。 、/ 輸入視訊圖框予，該視訊編碼器40 _ 作業包括習知的=碼（於步驟23G中）。該框間編石 必須於該框間編::後雙向預测。當1^框内編碼作業 碼器40產生介於該現日±執1丁（於步驟220中），該視訊編 考視訊圖框區塊間的請框之每-區塊與相應之參 該視訊編碼器4〇判^所块差值（於步驟240中）。接著； 預設參考值（於步驟π。生的纟巴對誤差總合值是否超避 與該參考視訊圖框 ）右介於該輸入視訊圖框區场 “ 產生的絕對誤差總合值小於t| I24686l 2t ，士值時，該交換位元設定為r 〇」（於步驟28〇中），並將 該父換位元值插入相應區塊之編碼資料間隔中。亦即，跨 越針對該相應區塊之框内編碼作業。 5 所吝Ϊ介於該輸入視訊圖框區塊與該參考視訊圖框區塊間 5的絕對誤差總合值超過該參考值時，該交換位元設 二為匕於步驟26〇中），並正常的將相應區塊予以框内 3 (於步驟270中）。此時，並將該交換位元值插入相應 &塊之編碼資料間隔中。 · 已4Ϊ:述之方法，當所有組成該輪入視訊圖框之區塊 期==框内編碼，該視訊編碼器4°於該框間編碼 』間再謂該輸人航_予赌_碼。 内編碼時㈣透過前述的方法將該輪人視㈣〜: 碼。透過前述之方法所峰忐 圖忙予以編 圖中。 斤生成之圖框陣列係顯示於第6與7 請參閱第6圖，第—艇&τ p 與娜次第的排列於該!圖框排:於前端，ΡWhen the information is encoded, the innovative inter-frame coding frame can have higher compression efficiency than the existing inter-frame coding frame. THIRD EMBODIMENT A third embodiment of the present invention provides a method of applying the aforementioned Pi frame and switching bits for increasing compression efficiency. The third embodiment of the present invention will be described in detail below with the accompanying drawings. 4 is a flow chart showing a video encoding process according to an embodiment of the present invention; and FIG. 5 is a flow chart showing a flow for decoding a view frame block according to an embodiment of the present invention; The 6 and 7 plans are illustrative of the arrangement of the frames generated by the video encoder 40 in accordance with the present invention. Referring to Figure 4, the video encoder 40 will be wheeled into the video frame for encoding in the first frame during the video encoding (in step 200). When the view frame is encoded as described above, the in-frame encoded video frame is stored in the reference frame memory unit 70 as a reference video frame (in the cow 210). And / input the video frame to, the video encoder 40_ job includes a conventional = code (in step 23G). The inter-frame masonry must be programmed between the boxes: post-bidirectional prediction. When the coded work code 40 of the frame is generated between the current date and the time (in step 220), each block of the frame between the video editing video frame blocks and the corresponding parameter The video encoder 4 determines the block difference (in step 240). Then; preset reference value (in step π. whether the raw sum of the error is over-avoided and the reference video frame) is right between the input video frame field "the absolute error sum is less than t | I24686l 2t, when the value is set, the swap bit is set to r 〇" (in step 28), and the parent transposition value is inserted into the coded data interval of the corresponding block. That is, the in-frame coding job for the corresponding block is crossed. 5 If the absolute error sum of the input video frame block and the reference video frame block exceeds the reference value, the exchange bit is set to be in step 26) And the corresponding block is normally placed in the box 3 (in step 270). At this point, the swap bit value is inserted into the encoded data interval of the corresponding & block. · 4 已: The method described, when all the blocks that make up the video frame are ===frame code, the video encoder 4° is coded between the frames, then the flight _ _ _ code. During the inner coding (4), the round person is regarded as (four) ~: code by the aforementioned method. Through the above methods, the peak map is busy being compiled. The array of frames generated by the jin is shown in the 6th and 7th. Please refer to Figure 6, the first row of the boat & τ p and Nadi is arranged in the frame: at the front end, Ρ
發明所生成之ΡΙ圖框取代被排列於該圖框透過本 圖框。該PI圖框係為依據 J中的另-I 請參閱第7圖，如圖所示已將3^向預測圖框。 框陣列中間部分之1圖框係依據時間間隔或二;f入該圖 生成:二Γ生成之1圖框係用作為參：:變化所新 尋介於先麵存的參相框與料作業時，搜 框區塊之區塊差異， ‘視讯圖框間具有圖 以框— 16 1246861 / 94. 6. 21 ^ =碼之視訊編物本發明之創新視訊編碼可增加壓 用以將透過前述視訊編碼器4q =之!簡以㈣接著區塊的方式描述於第 J閱第5 ® ’當框⑽碼視訊圖框之每— 二=訊解碼器6°判斷包含於重製或傳送編碼資料:: 換位元值為「0」時，該交換位元值「Q」表示乂 ，二已Λ對相應區塊跨越該編碼作業。該視訊料哭^ 該交換位元值為。」時，該視訊解：^ 相舰塊之編碼資料流予以解瑪（於步驟傷中）。 若所有視訊圖框中的區塊均透過前 以解碼之流程。依據前述的視鄭馬方 六^一碼益60依據透過該視訊編碼器40所插入之 ί St將相應區塊的編碼資料流予以解碼。可替代 :=::: =圖框區塊心 予以解碼。 予场碼’俾使視訊圖框可正常的 訊圖2述内容可明顯了解到，相較於習知用以將全部视 車^ iH ㈣編碼之魏糾方法本發明可增加魏效 兒至，、的&塊亚於该框内編碼作業必須予以執 17 1246861 行時僅將具有圖框差異之區塊予以内部編碼。 老：卜難本發明可依據時間間隔或環境變化自動更新參 :二Γ避免因時間間隔或環境變化所致生的壓縮效 質的降低。再者，本發明可應用於透過固定 优札I視糸統中以及如視訊會議系 =視2話等用以編碼與傳送具有小圖 框的系統中。 儘官用以說明用途知本發明較佳實施例已揭露於前， 熟習該項技彳㈣可时到在不㈣本發喊_情況下予 以不同的變化、增加或#效替換。據此，本發明並不限定 於前述之實施例，本發明及其等效之關將透過中請專利 範圍予以定義。 【圖式簡單說明】 第1圖係用以顯示習知視訊編碼器之方塊圖； 第2圖係用以顯示透過第！圖所示的視訊編碼器所生成之 圖框排列的一例示； 第3圖係用以顯示依據本發明之—實施例的視訊編碼/解 碼裝置之方塊圖， 第4圖係用以顯不依據本發明之_實施例之視訊編碼流程 之流程圖； 第5圖係用以顯不依據本發明之—實施例用以將視訊圖框 區塊解碼之流程的流程圖；以及 第6與7 以_輯過依據本發明之視訊編碼器所生 成的圖框排列例示。 18 1246861 / ·" (. ' - 【主要元件符號說明】 1 〇輸入緩衝器 12離散餘弦轉換 13量化器 14去量化器量化器 15反向離散餘弦轉換 17圖框記憶單元 18動態補償. 19動態估計器 20可變長度編碼 21輸出缓衝器 4 0視訊編碼器 50視訊流儲存單元 60視訊解碼器 70參考圖框記憶單元The frame generated by the invention is replaced by the frame arranged in the frame. The PI frame is based on the other -I in J. Please refer to Figure 7, as shown in the figure. The frame of the middle part of the frame array is based on the time interval or two; f is generated into the figure: the frame generated by the second generation is used as a parameter: the change is newly found in the first frame of the reference frame and the material operation , the difference between the blocks in the search box, 'the video frame has a picture frame — 16 1246861 / 94. 6. 21 ^ = code video editing The innovative video coding of the present invention can be increased to pass the video Encoder 4q =! Simplified in (4) followed by the block description in the Jth reading 5 ® 'in the frame (10) code video frame - two = decoder = 6 ° judgement included in the reproduction or transmission of coded data: : When the transposition value is "0", the exchange bit value "Q" indicates 乂, and the second block has crossed the coding operation for the corresponding block. The video material is crying ^ The exchange bit value is. At the time, the video solution: ^ The coded data stream of the phase ship block is solved (in the case of a step injury). If all the blocks in the video frame pass through the previous decoding process. According to the foregoing, Zheng Mafang Liuyiyiyi 60 decodes the encoded data stream of the corresponding block according to the ί St inserted through the video encoder 40. Alternate :=::: = The block block is decoded. The field code '俾 makes the video frame normal. The content of the picture 2 can be clearly understood. Compared with the well-known method for encoding all the cars ^ iH (four), the invention can increase the effect of Wei Xiaoer to The & block is only required to internally encode the block with frame differences when the in-frame encoding job must be executed on line 17 1246861. Old: Bu Dian This invention can automatically update the parameters according to time intervals or environmental changes: Secondly, to avoid the reduction of compression effect caused by time interval or environmental changes. Furthermore, the present invention can be applied to a system for encoding and transmitting a small frame through a fixed system and a video conferencing system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention has been disclosed in the prior art, and it is known that the technique (4) can be replaced with a different change, increase, or effect in the case of no (4) present. Accordingly, the invention is not limited to the foregoing embodiments, and the invention and its equivalents are defined by the scope of the patent application. [Simple diagram of the diagram] Figure 1 is used to display the block diagram of the conventional video encoder; Figure 2 is used to display the passage! An example of a frame arrangement generated by the video encoder shown in the figure; FIG. 3 is a block diagram showing a video encoding/decoding apparatus according to an embodiment of the present invention, and FIG. 4 is for displaying the basis A flowchart of a video encoding process of the embodiment of the present invention; FIG. 5 is a flowchart for illustrating a process for decoding a video frame block according to an embodiment of the present invention; and FIGS. 6 and 7 _ An illustration of the arrangement of the frames generated by the video encoder in accordance with the present invention. 18 1246861 / ·" (. ' - [Main component symbol description] 1 〇 input buffer 12 discrete cosine transform 13 quantizer 14 dequantizer quantizer 15 inverse discrete cosine transform 17 frame memory unit 18 dynamic compensation. 19 Dynamic estimator 20 variable length coding 21 output buffer 40 video encoder 50 video stream storage unit 60 video decoder 70 reference frame memory unit