1323133 九、發明說明： 【發明所屬之技術領域】 本發明大體上關於資料及影像編碼。更特定令 ^ 可應用本發明作為-用以提供資料或影像資訊或 訊之保護祕純正錄，諸如藉自在料絲像貝 應用層處應用一錯誤校正方案。 、、焉之 【先前技術】 包括音頻及影像資料之交換的多媒體通作在吾 生活中扮演著重要的角色。此資料交換可發生於:日常 線網路或媒體。例如，即時影像服務可在一多:或無 影像#料。然而，某些傳; 其係無線者）可能由於各_素(諸如衰料道尤 產生之資料損失、信號雜訊或信號干擾，而對 = 確或適當之資料傳輸造成挑戰。例如，-「 、&供精 劣化傳輸資料且在被發送之資料中產生無法預測可 業經提出某些機制以改進傳輸資料的精確二: 訊r及其他不符合需求之干擾。例如’正向錯誤 方案將冗餘位元引入在發送端的欲發送資料中，=㈣C) 在接收端處之解碼器使賴等冗餘位元以在傳輪 t 發t之錯誤。FEC的錯誤校正能力可取決於嵌入二 =冗餘總數。因此，錯誤校正能力可取決於可資料 輸之位元速率，且通道干㈣可㈣顯地㈣發送的貝=傳 純位元流以雜通料關料極為敏 感因為貝訊的任何損失可影響重建或已解螞之影像資 681954 6 ° 已去〇 減+ 之錯誤回復（error-resilience)方法使用不同方法來 】錯誤。例如，一些方法集中於使用空間或時間冗 爾从減少次几 碼位-士 化巨集區塊(ΜΒ)的錯誤。其他方法則對於已編 技術元=之標碩資訊提供資料保護，諸如HEC(標頭延伸碼） 括不相或封包層複製某些標頭資訊。而某些技術（包 之不同# 誤保護(UEP)技術)可提供賴# _分離區段 位準1確保資料可適#地解碼，祕提供可接受 旦貝之影像資訊。例如，可提供更可靠之子通道、更強的 FEC碼、或更頻繁的再傳輸。然而’現存網路、傳輸通作 協定以及傳輸系統可能不场UEP或某些錯_少技/ 因此，該等技術可能具有較低之應用價值或可能明顯地增 加應用成本。 由於現存方法或系統之某些限制，故需求一種改進資 料或影像的編碼方法與系統’可保護資料或影像或減少傳 輸錯誤影響之影像編碼方法或系統。 【發明内容】 參 符合本發明之辜巳例可提供-種影像編碼方法。該方法 4包括：接收〆組:於傳輸之影像資料；自該影像資料識 別至少二分區’ 一第一分區含有致能解碼該影像資料之解 瑪資訊的至少〜P刀，且一第二分區含有指示影像内容之 資訊的呈少…部分；編碼該第一分區之至少-部分於一系 統之〆應用層H區保護資訊’該分區保護資訊允 許當〆錯誤發生在該第—分區資料中時回復第_分區資 料；及發送該第-分區、該分區保護資訊、及該第二分區。 681954 7 1323133 符合本發明之範例亦可提供一執行影像編碼之系統。 該系統可包括：一輸入介面，其係用於接收一組用於傳輸 之影像資料；一資料處理裝置，其係用以自該影像資料識 別至少二分區，一第一分區含有致能解碼該影像資料之解 碼資訊的至少一部分，且一第二分區含有指示影像内容之 資訊的至少一部分；一編碼裝置，其係能編碼該第一分區 之至少一部分於一系統之一應用層，以產生分區保護資 訊，該分區保護資訊允許當該第一分區資料發生一錯誤時 • 回復第一分區資料；及一傳輸介面，其係用於發送該第一 分區、該分區保護資訊、及該第二分區。 符合本發明之範例亦可提供一電腦系統包括一記憶體 以儲存指令以及一處理器設定為處理指令。該處理器特別 係處理指令以進行步驟。包括：接收一組用於傳輸之資料； 自該資料識別至少二分區，一第一分區含有致能解碼該資 料之解碼資訊的至少一部分，且一第二分區含有指示該資 料内容之資訊的至少一部分；編碼該第一分區之至少一部 B 分於一系統之一應用層’以產生分區保護資訊，該分區保 護資訊允許當該第一分區資料發生一錯誤時回復該第一分 區資料；及發送該第一分區、該分區保護資訊、及該第二 分區。 當併同附圖閱覽以上詳細說明時，即可更佳暸解本發 明之此等及其他元件。 【實施方式】 現將參照本發明之具體實施例，其範例顯示於附圖中。 681954 8 1323133 符合本發明之範例提供用於編碼資料(諸如影像資料) 之方法及系統，其可對付在有線或無線網路不符合需求之 通道效應或接收資料的劣化。在一範例中’可在一系統之 應用層處應用一錯誤校正或資料保護技術，例如在一電腦 系統、無線影像系統或影像傳輸系統上執行應用程式。在 一些範例中，該等技術可應用在一應用層處，無須修改發 送多媒體或影像資料或位元串流(bitstream)的標準步驟或 語法。例如，一含有影像或其他資訊的編碼位元流於應用 層可透過一無線網路發送，且提供可接受或適當的資料， 用於在接收系統處重建該資訊或回復被錯誤影響的資 訊。在一些範例中，可應用一錯誤校正碼以保護一段影像、 音頻或多媒體資訊的標頭資訊。例如，可將錯誤校正碼嵌 入含有影像、音頻或多媒體資料之原始資料流中，其係藉 由將其置於一分離區段中，諸如在MPEG_4(動晝專家群 組’規格4)標準下提供之「user—defined—data」攔位，或藉 由使用資料隱藏技術。在一些範例中，在一應用層處應用 錯誤校正編碼可避免在保護欲發送之資料時需要一複雜 、統° …、 用於傳輸之多媒體（包括影像、音頻或其他資訊）可依 各種方式編碼以壓縮資料或提供適當精確度用於傳輸。例 如，影像編碼可使用一可變長度碼(VLC)方案以壓縮影像資 忒。然而’若一傳輸錯誤在一壓縮影像位元流中劣化一 符號’一在接收系統處的解碼器可能無法識別正確符號。 且在無法識別之符號後的資料可能由於在一群VLC符號中 681954 9 1323133 之相互相依性亦變得不可用。在一些範例中，若一錯誤發 生在指示解碼一群資料之位元流的標頭或一區段中時，其 影響將因為妨礙有關該標頭或區段之整個位元流的解碼而 明顯。 在一些範例中，影像編碼可針對重建發送資料可能需 要的位元流之標頭資訊或某些區段強調錯誤減少。例如， 可應用資料分區以將一影像位元流分成二個以上之分區， 例如具有二分區，其中一第一分區含有用於各巨集區塊之 編碼模式資訊，連同INTRA區塊之DC係數或INTER區 塊的運動向量，且一第二分區含有剩餘資料。例如，可應 用更多努力供保護解碼影像圖框必需之資訊。圖1顯示將 資料分區及不相等錯誤保護應用於分離分區的範例。參考 圖1，三種位準之重要性，如繪示於應用層之位準105、110 以及115可指派予用於傳輸之一或多個影像位元流中的不 同類型之資料，且FEC冗餘的不同位準、不同通道、不同 ARQ(自動再傳輸請求；Automatic Retransmission Request) 值或該等之結合亦可個別地應用於該等不同類型之資料。 例如，方塊120、125以及130繪示了在傳輸介面或系統位 準應用不同位準之保護或不同位準之系統需求以保護具有 不同位準之重要性的範例。 然而，取決於如何實施該技術，某些實施可能需要在 現存系統、傳輸協定或所需位元速率中明顯改變。例如， 如圖1所顯示，實行該方法所需的一特定設計網路。且一 些實施之有效性可取決於最新通道資訊之可用性。在一些 681954 10 1323133 範例中，小量的冗餘或資料保護資訊可併入作為用於傳輸 之原始資訊的一部分，並且系統提供者可能視受保護資料 ' 為普通位元流的一部分。但辨識該受保護資料之一接收系 統可能使用此資訊來校正或回復某資訊，諸如已發送資料 的影像標頭資訊或其他部分。 本發明的範例可包括保護或編碼影像、音頻、多媒體 或其他資料之方法。圖2a顯示在符合本發明之範例之一範 例流程圖。參照圖2a，例如，其一方法可包括在步驟140 • 接收一組資料，例如一組用於傳輸之影像資料或編碼影像 資料或一影像流。在步驟141可從資料或影像資料中識別 出二個以上的分區。例如，一第一分區可含有解碼資訊之 至少一部分，其致能解碼該資料（例如影像或其他編碼資 料）。在影像流之情況下，第一分區可含有一影像流之標頭 或該標頭的一部分。一第二分區可含有該資料本身的内 容，例如指示影像内容之資訊或此資訊的至少一部分。 為保護第一分區，在步驟142可將該第一分區或其至 ® 少一部分編碼於系統的應用層以產生分區保護資訊，當一 錯誤發生在該第一分區資料中時，其可允許第一分區資料 的回復。例如，若在一影像流的標頭中發生某些錯誤，該 分區保護資訊可允許某些或所有受錯誤影響之標頭資訊回 復或重新配置。接著可發送第一分區、分區保護資訊及第 二分區，諸如在步驟143透過有線或無線網路發送到一或 多個資料接收系統。在一範例中’可結合第一分區、分區 保護資料以及第二分區為一保護位元流，其可維持原有影 681954 11 1323133 像或資料的語法。 至少: 個以上分區及編碼第-分區之 生此應St辨争輸系㈣應用層發 生。此應用可避免修改傳輸標 化影像系統的設計。在另一範合丨中。《之南要’且簡 匚統的媒體存=::= 類生：始資:二==;第二分區可依- 法中姑W也丄輸堵如原始影像資料的語 統仍能接或認識上述編碼機制之_ 雜校正碼作為該分區保護資=有用 :二Τ:包括對於影像資料之標頭;=!錯 ^^"(Hamming) 、x正4。在編碼1像位域之範例中，第 分區保護資訊及第二分區可在—個影像位元流^ 流：資之:rrrfm之影像位元 瑪以保護此資訊或其 應用=處實打此過程可避免需要祕m在錯誤 编瑪< ’分區保護資減因為錯誤校正編碼產生之冗餘 可與影像資谢綱普物fw發㈣^在 681954 12 1323133 應用層處應用錯誤校正編碼之範例。參考圖^，諸如 =:(:;象:件，面’之資料或影像巨集區塊_的標 ’、，可用正向錯誤校正(FEC)碼保護。並且該方 法可僅使用單-運輪及實體通道而無_外之_或需要 最佳直質傳輪之通道。在一些範例中，此技術可減少或 避免需要調整-些目子，例如FEC編碼㈣ ’、 (自動再發送請求）的數目，以增加重要資訊的ί正 可能性。1323133 IX. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention generally relates to data and image coding. More specifically, the invention can be applied as a means of providing a confidentiality of information or video information or information, such as by applying an error correction scheme at the application layer of the wire. [Previous Technology] Multimedia communication, including the exchange of audio and video materials, plays an important role in my life. This exchange of information can occur on: daily network or media. For example, the instant image service can be more than one: or no image # material. However, some transmissions; which are wireless) may pose challenges to the transmission of data, whether or not, due to data loss, signal noise or signal interference. For example, - , & fine degradation of transmission data and unpredictable in the transmitted data can be proposed to improve the accuracy of the transmission of data: interference and other non-conformance interference. For example, 'forward error scheme will be redundant The remaining bits are introduced in the data to be sent at the transmitting end, = (4) C) The decoder at the receiving end causes the redundant bits to be erroneously transmitted in the transmission t. The error correction capability of the FEC may depend on the embedded two = The total number of redundancy. Therefore, the error correction capability can depend on the bit rate at which data can be transmitted, and the channel is dry (4) can be (4) explicit (4) sent by the shell = pass pure bit stream is extremely sensitive to the miscellaneous material because Beixin Any loss can affect the reconstruction or resolution of the image of the 681954 6 ° has been derogated + error-resilience method using different methods to error. For example, some methods focus on the use of space or time Reduce the error of the next few code-scientific block (ΜΒ). Other methods provide data protection for the standard information of the programmed technology element, such as HEC (header extension code), including phase or packet layer replication. Some header information. Some technologies (packaged different #false protection (UEP) technology) can provide Lai _ separation segment level 1 to ensure that the data can be decoded, the secret provides acceptable image information. For example, a more reliable subchannel, a stronger FEC code, or a more frequent retransmission may be provided. However, the existing network, the transport protocol, and the transmission system may not be UEP or some error/small/ Techniques may have lower application value or may significantly increase application costs. Due to certain limitations of existing methods or systems, there is a need for an improved coding method and system for data or images that can protect data or images or reduce the effects of transmission errors. Image coding method or system. SUMMARY OF THE INVENTION An image coding method can be provided in accordance with an example of the present invention. The method 4 includes: receiving a group of images: image data transmitted; Data identifying at least two partitions 'a first partition containing at least a P-knife capable of decoding the decoded information of the image data, and a second partition containing a portion indicating information of the image content; encoding the first partition At least - part of the system application layer H zone protection information 'this partition protection information allows to reply to the _ partition data when the error occurs in the first partition data; and send the first partition, the partition protection information, And the second partition. 681954 7 1323133 An example consistent with the present invention may also provide a system for performing image coding. The system may include: an input interface for receiving a set of image data for transmission; The device is configured to identify at least two partitions from the image data, a first partition containing at least a portion of decoding information capable of decoding the image data, and a second partition containing at least a portion of information indicating the content of the image; The device is capable of encoding at least a portion of the first partition in an application layer of a system to generate partition protection information, the Protection of information allows data partition when the first error occurs • a reply to the first partition data; and a transmission interface, its system for sending the first partition, the partition information protection, and the second partition. An example consistent with the present invention may also provide a computer system including a memory for storing instructions and a processor for processing instructions. The processor specifically processes the instructions to perform the steps. The method includes: receiving a set of data for transmission; identifying at least two partitions from the data, a first partition containing at least a portion of decoding information capable of decoding the data, and a second partition containing at least information indicating the content of the data Part of: encoding at least one portion B of the first partition into an application layer of a system to generate partition protection information, the partition protection information allowing the first partition data to be replied when an error occurs in the first partition data; and Sending the first partition, the partition protection information, and the second partition. These and other elements of the present invention will be better understood from the following detailed description. [Embodiment] Reference will now be made to the specific embodiments of the invention, 681954 8 1323133 An example consistent with the present invention provides a method and system for encoding data, such as image data, that can address channel effects or degradation of received data that are not in compliance with a wired or wireless network. In one example, an error correction or data protection technique can be applied at the application layer of a system, such as executing an application on a computer system, a wireless imaging system, or an image transmission system. In some examples, the techniques can be applied at an application layer without the need to modify standard steps or syntax for transmitting multimedia or video material or bitstreams. For example, a coded bitstream containing images or other information can be sent over the wireless network over the application layer and provide acceptable or appropriate data for reestablishing the information at the receiving system or for replying to the information that was affected by the error. In some examples, an error correction code can be applied to protect the header information of a piece of image, audio or multimedia information. For example, the error correction code can be embedded in the original data stream containing the image, audio or multimedia material by placing it in a separate section, such as under the MPEG_4 (Moving Expert Group 'Specification 4') standard. Provide a "user-defined-data" block, or by using data hiding techniques. In some examples, applying error correction coding at an application layer avoids the need for a complex, unified multimedia device (including video, audio, or other information) that can be encoded in various ways when protecting the data to be transmitted. Compress the data or provide the appropriate accuracy for transmission. For example, image encoding can use a variable length code (VLC) scheme to compress image assets. However, if a transmission error degrades a symbol in a compressed image bitstream, a decoder at the receiving system may not be able to recognize the correct symbol. And the data behind the unrecognizable symbols may also become unavailable due to the interdependence of 681954 9 1323133 in a group of VLC symbols. In some examples, if an error occurs in a header or a section indicating the bitstream of a group of data, the effect will be apparent by obstructing the decoding of the entire bitstream for that header or section. In some examples, image encoding may emphasize error reduction for header information or certain segments of the bitstream that may be needed to reconstruct the transmitted data. For example, data partitioning may be applied to divide an image bit stream into more than two partitions, for example, having two partitions, wherein a first partition contains encoding mode information for each macroblock, together with DC coefficients of the INTRA block. Or the motion vector of the INTER block, and a second partition contains the remaining data. For example, more effort can be applied to protect the information necessary to decode the image frame. Figure 1 shows an example of applying data partitioning and unequal error protection to separate partitions. Referring to Figure 1, the importance of the three levels, as shown in the application layer levels 105, 110, and 115, can be assigned to transmit different types of data in one or more image bitstreams, and FEC redundancy The different levels, different channels, different ARQ (Automatic Retransmission Request) values or combinations of these may also be applied individually to the different types of data. For example, blocks 120, 125, and 130 illustrate examples of applying different levels of protection or different levels of system requirements at the transmission interface or system level to protect the importance of having different levels. However, depending on how the technology is implemented, some implementations may require significant changes in existing systems, transport protocols, or required bit rates. For example, as shown in Figure 1, a particular design network is required to implement the method. And the effectiveness of some implementations may depend on the availability of the latest channel information. In some examples of 681954 10 1323133, a small amount of redundancy or data protection information may be incorporated as part of the original information for transmission, and the system provider may consider the protected material 'as part of the normal bit stream. However, the receiving system that identifies the protected data may use this information to correct or reply to certain information, such as image header information or other parts of the transmitted material. Examples of the invention may include methods of protecting or encoding images, audio, multimedia or other materials. Figure 2a shows a flow chart in an example of an example consistent with the present invention. Referring to Figure 2a, for example, a method can include, at step 140, receiving a set of data, such as a set of image data or encoded image data or a stream of images for transmission. In step 141, more than two partitions can be identified from the data or image data. For example, a first partition may contain at least a portion of the decoded information that enables decoding of the material (e. g., images or other encoded material). In the case of a video stream, the first partition may contain a header of an image stream or a portion of the header. A second partition may contain the content of the material itself, such as information indicating the content of the image or at least a portion of this information. To protect the first partition, the first partition or a portion thereof to the application layer of the system may be encoded in step 142 to generate partition protection information, which may be allowed when an error occurs in the first partition data. A reply to a partition data. For example, if some error occurs in the header of an image stream, the partition protection information may allow some or all of the header information affected by the error to be replied or reconfigured. The first partition, the partition protection information, and the second partition can then be sent, such as in step 143, via a wired or wireless network to one or more data receiving systems. In one example, the first partition, the partition protection data, and the second partition may be combined as a protected bit stream that maintains the syntax of the original image 681954 11 1323133 image or material. At least: more than one partition and the coded first-partition should be the St-distribution system (4) application layer. This application avoids the modification of the design of the transmitted standard image system. In another category. "Southern" and simple mediation =::= class students: capital: two ==; the second zone can be based on - the law of the aunt, but also the language of the original image data can still be connected Or recognize the above-mentioned coding mechanism _ miscellaneous correction code as the partition protection = useful: second: including the header for the image data; =! wrong ^ ^ " (Hamming), x positive 4. In the example of encoding a 1-bit field, the first partition protection information and the second partition can be in the image stream stream: the image bit: rrrfm image bit to protect the information or its application = The process can avoid the need for the secret m in the error encoding < 'partition protection reduction' because the redundancy generated by the error correction coding can be used with the image of the Xiepu object fw (four) ^ at the application layer of 681954 12 1323133 application error correction coding. Referring to Figure ^, such as =: (:; like: piece, face 'data or image macro block _ mark', can be protected by forward error correction (FEC) code. And the method can only use single-transport Round and physical channels without _ or _ or the channel that requires the best direct quality. In some examples, this technique can reduce or avoid the need to adjust - some objects, such as FEC encoding (4) ', (automatic resend request The number of times to increase the positive likelihood of important information.
範例中，第—純心可僅包括影像標頭資訊 3 71因此，由於編碼過程產生的額外分區保護資 =受到限制。在將漢日㈣用作應用層中之錯誤校正方法的 =二，僅需要lkps之額外位元速率’用於特定之影像流。 ^且在—些範例中，由於漢明碼之特徵，編碼操作不改變 =影響原始標頭資訊。在各種範例中，分區保護資訊可易 ^併入原始位元流中而無須修改現存鮮之語法，包括在 gMPEG，晝專家群，規格4)等標準下㈣像編碼/ 在-些㈣中’分區保護資訊、小量冗餘可包括在使 用數位浮水印或資料隱藏技術的位元流_。 同時’可將分區保護資訊置於影像流之某些爛位或區 心中，例如依據MPECM標準提供之「讀_defined她」 或其他攔位中。在-些範例中’因為（例如藉由漢明碼)編 碼第-分區或其-部分不影響第―分區或其本身的標頭資 訊’該分d保護資訊可包括在依循其原始語法之影像資料 681954 13 丄323133 或衫像位元流中。在一些範例中，可包括分區保護資訊作 =第二分區145b或影像内容之一部分，包括作為替代影像 二貝料之較不或最不重要位元的某些資料。在一些其他的範 例中，可將分區保護資訊附加至第二分區145b或影像資料 的末端。在其中遵循原始語法或類似語法之範例中，一不 支援所提出錯誤校正技術（例如應用層錯誤校正）之解碼器 仍可解碼一已編碼的位元流。如上所述，方塊145a與145b 可包括第一分區、分區保護資訊以及第二分區，可在應用 150結合或編譯為一保護位元流。可使用不需訂做設計之 通用或一般網路傳輸位元流。並且可應用如方塊16〇所顯 不’在傳輸介面或系統位準具有平均或一般保護的系統需 求。 範例性模擬已使用上揭方法執行。在一些範例中，雖 然應用該編碼操作可能稍微增加若干百分比之資料總數 (例如約1%或更少），但傳輸速率可明顯地減少，尤其係在 無線應用中，因為該傳輸需要較少的冗餘或保護。例如從 系統觀點而言，應用層的資訊玎藉由在實際傳輸發生前使 用FEC編碼或其他編碼而進一少被保護。例如，在傳輸後’ 可根據影像解碼器可忍受之位元錯誤率（BER)’提供可接受 影像資料之FEC冗餘位準或通道編碼的編碼率。不嚴格的 BER需求可導致通道編碼所需冗餘位準之減少。因為在一 組接收到之影像資料中的某些資料係由分區保護資訊所保 護’故使BER需求較不嚴格、通道編碼之編碼率的減少或 二者均變得可行。 681954 14 1323133 例如，對於一在】0·3 嚴格的酬處操作的影像解二 對較不 化’其導致在接收系統處嚴重降級之圖片晝質：：成劣 後反通常會需要1〇、上之，提4= 一之，應用以上所示技術可為諸如用沐又 解碼程序中之標頭資訊的重要資訊提供已改進的回復影像 3』不需要滿;1BER之不同位準的通道編^ 。圖 =例。參考圖3,其顯示對應於三個不同編碼;::: -。對於約5分貝之雜訊比’滿足1〇-3的咖-= 的編碼率或12.5%之冗餘率。滿足⑽之咖 =/9 碼率或50%之冗餘率。在一些範例令，所建議 的， 傳輸位元速率或需要提供適當資料傳輪之 範例性模擬亦係在可能發生於提供即時影像 例性情節下實施。第-情節可發生在廣播服務中其^ 通道碼編碼器或影像編碼ϋ可能無法從—或多個接收 得到回授’以調適其編碼過程。因此，錯誤可能跟著=門 而傳播，直到在接收系統處之影像解碼器接收到一^ INTKA d塊。第二情節考慮用於-即時影像服務之回授通 道。例如，當-些資料或圖片的傳輪由於某些資訊(例如在 圖片標頭）中錯狀賴錢效時，可通知1道碼編碼器 或影像編喝ϋ。因此…通道碼編㈣或影像編瑪器可使 用此等通知作為回授資訊且嘗試減少或停止錯誤傳播。 在一範例中，可將漢明碼用作一錯誤校正碼以保蠖一 681954 15 1323133In the example, the first-pure heart can only include the image header information 3 71 Therefore, the additional partition protection generated by the encoding process is limited. In the use of Han (4) as the error correction method in the application layer = 2, only the extra bit rate of lkps is required for a specific image stream. ^ And in some examples, due to the characteristics of Hamming code, the encoding operation does not change = affect the original header information. In various examples, the partition protection information can be easily incorporated into the original bit stream without modifying the existing syntax, including gMPEG, 昼 expert group, specification 4), etc. (4) Image coding / in - (4) Partition protection information, small redundancy can be included in the bit stream _ using digital watermarking or data hiding technology. At the same time, the partition protection information can be placed in some rotten areas or areas of the video stream, such as "read_defined her" or other intercepts provided by the MPECM standard. In some examples, 'because (for example, by Hamming code) the first-partition or its part does not affect the first-partition or its own header information'. The d-protection information can be included in the image data according to its original grammar. 681954 13 丄 323133 or shirt like a bit stream. In some examples, the partition protection information may be included as part of the second partition 145b or image content, including some of the lesser or least significant bits of the alternate image. In some other examples, partition protection information may be appended to the second partition 145b or the end of the image data. In an example in which the original grammar or similar grammar is followed, a decoder that does not support the proposed error correction technique (e.g., application layer error correction) can still decode an encoded bit stream. As discussed above, blocks 145a and 145b can include a first partition, partition protection information, and a second partition that can be combined or compiled into a protected bitstream at application 150. A general purpose or general network transport bit stream that does not require a custom design can be used. It is also possible to apply a system requirement that has an average or general protection at the transmission interface or system level as indicated by block 16 。. An exemplary simulation has been performed using the above method. In some examples, although the application of the encoding operation may increase the number of data by a certain percentage (eg, about 1% or less), the transmission rate may be significantly reduced, especially in wireless applications, since the transmission requires less Redundancy or protection. For example, from a system perspective, the application layer information is less protected by using FEC encoding or other encoding before the actual transmission occurs. For example, the FEC redundancy level or channel coding rate of acceptable image data can be provided after transmission based on the bit error rate (BER) that the image decoder can tolerate. Less stringent BER requirements can result in a reduction in the redundancy levels required for channel coding. Because some of the information received in a group is protected by partition protection information, the BER requirements are less stringent, the channel coding rate is reduced, or both become feasible. 681954 14 1323133 For example, for a image that is strictly operated at 00.3, the two pairs of images are less likely to cause a picture that is seriously degraded at the receiving system:: After the inferiority, it usually takes 1〇, 4, one, the application of the above technology can provide improved reply image 3 for important information such as the use of the header information in the decoding process, no need to be full; 1BER different levels of channel editing ^ . Figure = example. Referring to Figure 3, the display corresponds to three different codes; :::: -. For a noise ratio of about 5 dB, 'the code rate of the coffee-= of 1〇-3 or the redundancy rate of 12.5%. Satisfy (10) coffee = / 9 code rate or 50% redundancy rate. In some example orders, the proposed transmission rate of the bit rate or the need to provide an appropriate data transfer is also implemented in an example scenario where it may occur to provide an instant image. The first episode may occur in the broadcast service where its channel code encoder or image coding may not be able to receive feedback from - or multiple receptions to adapt its encoding process. Therefore, the error may propagate with the = gate until the image decoder at the receiving system receives an INTKA d block. The second episode considers the feedback channel used for the instant image service. For example, when the data of a certain data or picture is misused due to some information (such as in the picture header), the code code encoder or image can be notified. Therefore, the channel code (4) or image coder can use these notifications as feedback information and try to reduce or stop the error propagation. In an example, the Hamming code can be used as an error correction code to protect a 681954 15 1323133
组資料的某些資訊，例如一 H.264影像位元流之標頭資 訊。該標頭資訊可包括一序列層標頭、一圖片層標頭、及 一片（slice)層標頭。在一範例中，用以保護此等標頭之冗餘 位元總數係約lkbps。錯誤可藉由在1〇·3之BEr處施加隨 機雜訊以劣化二範例性位元流而產生，其一具有標頭保護 而另一者不具有標頭保護。可設定一些準則用於該等範例 性模擬。例如，可藉由從最靠近之圖框複製資訊以取代已 劣化的圖框，並且在第一圖框中的標頭資訊沒有錯誤。若 干影像序列可用於測試該系統效能。各序列皆可含有一些 圖框，諸如300圖框且可編碼為μρ-ρ-ρ-ρ-ρ .，其意味著 該第一圖框係框内編碼（intra_c〇decj)而所有其他圖框皆係 框間編碼（inter-coded)。 通常 用；八有^頭保護之各種序列的各種pSNR值 係高於不具標頭保護之各種序列的該等值。 大體上，因為二鄰近圖框可能極不同，在具有一已劣 化標頭後使用鄰近圖框之資訊將會嚴重地劣化在快速移動 序列中重建的圖片°雖然在—初始圖框後之圖框中的一些 INTRA區塊可允許回復已劣化圖片以稱微地改進PSNR效 月b仁此等INTRA區塊仍無法補償在一已劣化標頭中之錯 ^並且料錯翻·傳_ 餘進 PSNR效能降級。 is 對於一具有回授的系統而 像解碣器某些資訊業已劣化且 對解碼器提供此資訊之潛時可 言’該回授通道可通知一影 不應用作參考資訊。可假設 花費二圖框的時間間隔。因 681954 16 1323133 此，任何之後的圖框可能具有一適當之參考並且可停止錯 誤傳播。 ' 在一些範例中，可提供一能執行資料編碼或影像編碼 之系統以實施上揭方法。例如，該系統可包括一輸入介面， 用於接收一組用於傳輸之影像資料。該系統亦可包括一用 於自影像資料識別至少二分區的資訊處理裝置。尤其一第 一分區可含有致能解碼該影像資料之解碼資訊的至少一部 分，且一第二分區可含有指示影像内容之資訊的至少一部 籲 分。該系統亦可包括一編碼裝置，其係能編碼該第一分區 之至少一部分，以產生分區保護資訊。該分區保護資訊允 許當一錯誤發生在該第一分區資料中時’回復第一分區資 料。此外，該系統可包括一傳輸介面，其係用於發送該第 一分區、該分區保護資訊、及該第二分區。 可以使用不同種類的系統實行前示之方法。例如，系 統可以是電腦、一處理器在多媒體傳輸或影像傳輸系統、 一專用硬體、一處理器執行一組韌體或軟體指令、一邏輯 ’組合等。 如上揭，用於編碼資料（例如影像資料）之方法可提供 保護致能解碼該資料（例如影像位元流的標頭）之解碼資訊 的至少一部分。在一範例中，可在一應用層處應用錯誤校 正編碼。在一易於錯誤的通道中，該等方法可避免或減少 標頭資訊錯誤’以在一些範例中允許已改進的影像解碼。 在一些範例中，該等方法可免於需要修改現存標準之語 法，並且可用各種技術在位元流中隱藏冗餘位元。此實施 681954 17 可提供與現存標準之相办 應用於任何通信網路或傳‘環可將該等方法之範例 本發明之範例的前揭 其並非欲毫無遺漏或 2係、供例不及說明之目的。 於以上的揭示内容，太發明為所揭露之精確形式。基 熟習此項技藝者而述範例之許多變化及修錦對於 文所附之申請專“園知的。本發明之範嘴僅由本 另外，在說明定義。 本發明之方法或製程表- 範例時，本說明書可將 :r法或製㈣範二由 ，人序，故該方法或製程 所&出之特疋的步驟 身為熟習本技藝者當會了 所述之特定步驟次序。 以，不應將本說明書=步驟次序亦是可行的。所 利範圍的限制。此外，介 特定步驟次序視為對申請專 專利範圍僅限制在以金=將有關本發明之方法的申請 項技藝者易於瞭解，；等次驟次序之實施’熟習此 於本發明之精神與範疇内。’、可加以改變，並且仍涵蓋 【圖式簡單說明】 前揭:而閲覽時’即可更佳瞭解本發明之 圖式裹圖綠有現屬 並不限於所緣之精_ 、，實知例。然應瞭解本發明 圖1顯干2 式及設備裝置。在各圖式中： 之範例/〜用貧料分區及不相等錯誤保護於分離分區 681954 18 1323133 圖2a顯示在符合本發明之範例之一範例流程圖。 圖2b顯示在符合本發明之範例中的一應用層處應用 錯誤校正編碼之範例。 圖3顯示需要滿足符合本發明之範例中的BER之不同 位準的通道編碼之不同編碼率的範例。Some information about the group data, such as the header information of a H.264 image bit stream. The header information can include a sequence layer header, a picture layer header, and a slice layer header. In one example, the total number of redundant bits used to protect these headers is approximately lkbps. Errors can be generated by applying random noise at the BEr of 1〇·3 to degrade the two exemplary bitstreams, one with header protection and the other without header protection. Some criteria can be set for these exemplary simulations. For example, the degraded frame can be replaced by copying the information from the closest frame, and the header information in the first frame is error free. Several image sequences can be used to test the performance of the system. Each sequence may contain some frames, such as a 300 frame and may be encoded as μρ-ρ-ρ-ρ-ρ . , which means that the first frame is intra-frame encoded (intra_c〇decj) and all other frames Both are inter-coded. Typically, the various pSNR values for the various sequences of the eight-head protection are higher than those for the various sequences without header protection. In general, because the two adjacent frames may be very different, using the information of the adjacent frames after having a degraded header will seriously degrade the reconstructed picture in the fast moving sequence. Although in the frame after the initial frame Some of the INTRA blocks may allow the reply to the degraded picture to be slightly improved. The INTRA block still cannot compensate for the error in a degraded header and the error is reversed. Performance degradation. Is for a system with feedback, like some information that the decoder has degraded and provides this information to the decoder. The feedback channel can notify a shadow that should not be used as reference information. It can be assumed that the time interval of the two frames is spent. As a result of 681954 16 1323133, any subsequent frames may have an appropriate reference and may stop error propagation. In some examples, a system capable of performing data encoding or image encoding can be provided to implement the method. For example, the system can include an input interface for receiving a set of image data for transmission. The system can also include an information processing device for identifying at least two partitions from the image data. In particular, a first partition may contain at least a portion of the decoded information that enables decoding of the image data, and a second partition may contain at least one portion of the information indicative of the image content. The system can also include an encoding device that encodes at least a portion of the first partition to generate partition protection information. The partition protection information allows the first partition data to be replied when an error occurs in the first partition data. Additionally, the system can include a transmission interface for transmitting the first partition, the partition protection information, and the second partition. The methods shown above can be implemented using different kinds of systems. For example, the system can be a computer, a processor in a multimedia transmission or video transmission system, a dedicated hardware, a processor executing a set of firmware or software instructions, a logical combination, and the like. As noted above, the method for encoding data (e. g., image material) can provide protection for at least a portion of the decoded information that enables decoding of the material (e.g., the header of the image bitstream). In an example, an error correction code can be applied at an application layer. In an error-prone channel, these methods can avoid or reduce header information errors' to allow for improved image decoding in some examples. In some examples, the methods are exempt from the need to modify the syntax of existing standards, and various techniques can be used to hide redundant bits in the bitstream. This implementation 681954 17 may provide for the application of any standard communication system or any of the methods of the present invention. The examples of the present invention are not intended to be exhaustive or to be omitted. The purpose. The above disclosure is too invented to be the precise form disclosed. Many variations and modifications of the examples described herein will be apparent to those skilled in the art. The scope of the present invention is defined solely by the present specification. The method or process table of the present invention - examples This specification may be: r method or system (4) fan two, human order, so the method or process of the steps and the special steps are familiar to the skilled person when the specific step sequence is described. This specification should not be construed as a step-by-step procedure. The scope of the specific steps is considered to be limited to the scope of the patent application only to the applicants of the method of the present invention. Understand the implementation of the sub-sequences 'experienced within the spirit and scope of the present invention.', can be changed, and still cover [simplified description of the schema] before the release: while reading, you can better understand this The invention is not limited to the essence of the invention. However, it should be understood that the present invention is shown in Fig. 1 and the device device. In each of the drawings: Poor material partition and unequal error protection Partition 681954 18 1323133 Figure 2a shows an example flow diagram in accordance with one example of the present invention. Figure 2b shows an example of applying error correction coding at an application layer in accordance with an example of the present invention. Figure 3 shows the need to meet the requirements of the present invention. Examples of different coding rates for channel coding of different levels of BER in the example.
681954 19681954 19