WO2012000379A1 - 一种提高视频解码图像质量的方法及解码器 - Google Patents

一种提高视频解码图像质量的方法及解码器 Download PDF

Info

Publication number
WO2012000379A1
WO2012000379A1 PCT/CN2011/075582 CN2011075582W WO2012000379A1 WO 2012000379 A1 WO2012000379 A1 WO 2012000379A1 CN 2011075582 W CN2011075582 W CN 2011075582W WO 2012000379 A1 WO2012000379 A1 WO 2012000379A1
Authority
WO
WIPO (PCT)
Prior art keywords
macroblock
abnormal
code stream
decoding
decoder
Prior art date
Application number
PCT/CN2011/075582
Other languages
English (en)
French (fr)
Inventor
梁立伟
王宁
范文杰
黄晓红
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2012000379A1 publication Critical patent/WO2012000379A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/174—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a slice, e.g. a line of blocks or a group of blocks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/85—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression
    • H04N19/89—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving methods or arrangements for detection of transmission errors at the decoder
    • H04N19/895—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving methods or arrangements for detection of transmission errors at the decoder in combination with error concealment

Definitions

  • the invention belongs to the field of video communication, and particularly relates to a method and a decoder for improving the quality of video decoded images.
  • H.263, H.264, and MPEG-4 (Moving Picture Experts Group) video coding standards have been specified in the H.324-M specification used in the current third generation mobile communication (3rd Generation, 3G) technology standard. These standards are based on motion-compensated video coding algorithms. Most of the video frames are highly correlated with the previous frames. Any frame error will be propagated and spread to subsequent frames, causing video quality.
  • 3G Third Generation
  • the loss of the code stream directly leads to the loss of the decoded information, which causes the decoding to be erroneous, and the error of the code stream will result in the error of the decoded information, which also affects the quality of the decoded video.
  • the code stream is lost, if the decoder does not find the occurrence of the packet loss in time, and the subsequent code stream packet is decoded as the current code stream, the consequences similar to the code stream error will occur.
  • the main phenomenon That is, serious color blocks, misaligned orphan blocks, etc. appear when decoding.
  • code stream errors have many different forms of expression, such as header errors, motion vector errors, and quantization coefficient errors.
  • Head errors can cause subsequent data to be decoded (including motion vectors and quantized coefficients).
  • Motion vector errors can cause motion compensation positions to be incorrect, resulting in isolated blocks.
  • An incorrect quantization factor can result in an incorrect brightness or chrominance value.
  • the abnormal macroblocks corresponding to the abnormal code stream are often connected or similar in position, and the abnormal macroblocks with similar positions may affect the decoding process of the normal macroblocks adjacent thereto, and the abnormal macroblock may also cause subsequent normal macroblocks. Decoding error.
  • the prior art typically determines a lag abnormal macroblock, shown in Figure 1, in the bitstream macroblock MB n erroneous macroblock is already in the decoding process, but this does not meet the macro block MB n decoders Judging the condition of the abnormal macroblock, which may cause the decoder to solve
  • the code reaches the macroblock MB n+m, an error is found, so that after the macroblock between MB n and MB n+n>1 is decoded, the image has a phenomenon of color block or misalignment of the isolated block.
  • the technical problem to be solved by the present invention is to provide a method and a decoder for improving the quality of a video decoded image, which effectively eliminates the phenomenon of color blocks or misaligned isolated blocks in a decoded image caused by packet loss or bit error, and improves the quality of the decoded image.
  • the present invention provides a method for improving the quality of a video decoded image, including: in the process of decoding a video image compressed data stream, the code stream is detected to be abnormally processed.
  • the foregoing method may further have the following features:
  • the step of determining the abnormal macroblock after detecting the code stream abnormality comprises: when the data loss condition is detected during decoding, the data packet in which the last macroblock obtained by the complete decoding is detected in the decoding process before the data loss is detected All macroblocks in the block are determined to be abnormal macroblocks.
  • the foregoing method may further have the following features:
  • the step of determining the abnormal macroblock after detecting the code stream abnormality comprises: when the syntax error is detected for the code stream in the decoding process, all the macroblocks in the data packet in which the syntax error macroblock is located are determined as the abnormal macroblock.
  • the foregoing method may further have the following features:
  • the detection of a syntax error means that one or more of the following conditions are detected:
  • the motion vector in the macroblock is out of the valid range, the variable length codeword is invalid, the value of the discrete cosine transform coefficient is outside the valid range, and the number of discrete cosine transform coefficients is more than valid.
  • the range, the quantized coefficient is out of the valid range, and the number of decoded macroblocks is out of the valid range.
  • the foregoing method may further have the following features:
  • the data packet refers to one frame of image data specified in the video compression standard, or a macroblock group, or a video packet, or a stripe.
  • the present invention further provides a decoder for improving video decoded image quality, the decoder comprising a code stream abnormality determining module and a covert processing module; and the code stream abnormality determining module is set as: In the process of decoding the video image compressed data stream, detecting a location of the constant macroblock is notified to the concealment processing module; the concealment processing module is configured to: at an abnormal macroblock position learned from the code stream abnormality determining module The data is concealed.
  • the above decoder may also have the following features:
  • the code stream abnormality determining module is configured to determine a decoding process before the time when the data loss is detected when the abnormal macroblock detects the data loss when the decoder performs decoding in the following manner. All macroblocks in the last packet in which the fully decoded macroblock is located are determined to be abnormal macroblocks.
  • the above decoder may also have the following features:
  • the code stream abnormality determining module is configured to determine, when the code stream abnormality is detected, to determine that the abnormal macroblock is in the process of decoding by the decoder, and when a syntax error is detected for the code stream, a macroblock in which a syntax error occurs is located. All macroblocks in the packet are determined to be abnormal macroblocks.
  • the above decoder may also have the following features:
  • the code stream abnormality determining module detects that the syntax error refers to detecting one or more of the following cases: the motion vector in the macro block exceeds the valid range, the variable length codeword is invalid, and the value of the discrete cosine transform coefficient exceeds the effective range, and the discrete The number of cosine transform coefficients exceeds the effective range, the quantized coefficients exceed the valid range, and the number of decoded macroblocks exceeds the valid range.
  • the above decoder may also have the following features:
  • the data packet refers to one frame of image data specified in the video compression standard, or a macroblock group, or a video packet, or a stripe.
  • the determined abnormal macro block is in units of data packets, thereby effectively eliminating color blocks or misaligned isolated blocks appearing in video image decoding caused by packet loss or bit error. phenomenon. Especially in the case of poor code stream quality, it effectively suppresses the adverse effects of abnormal macroblocks connected or similar in position on the decoding process of adjacent normal macroblocks and abnormal macroblocks. The adverse effect on the decoding process of subsequent normal macroblocks improves the overall viewing effect of the decoded video and improves the user experience.
  • FIG. 1 is a schematic diagram showing a hysteresis when an abnormal macroblock is judged in the prior art
  • FIG. 2 is a structural diagram of a decoder for improving video image quality in an embodiment of the present invention
  • FIG. 3 is a flowchart of a method for improving video decoded image quality in an embodiment of the present invention
  • FIG. 4 is a diagram showing a decoding example of an exemplary embodiment 1 of the embodiment of the present invention.
  • FIG. 5 is a diagram showing an example of decoding of an exemplary embodiment 2 in the embodiment of the present invention. Preferred embodiment of the invention
  • the decoder for improving the quality of the video decoded image includes a code stream abnormality determining module 20 and a covert processing module 21.
  • the code stream abnormality determining module 20 is arranged to decode the video image compressed data stream in units of the decoder and notify the covert processing module of the position of the abnormal macroblock.
  • the concealment processing module 21 is arranged to perform concealment processing on the data at the abnormal macroblock position learned from the code stream abnormality determining module.
  • the decoder can typically process the data loss condition, and the code stream abnormality determining module 20 is configured to: when the data loss is detected when the decoder performs decoding, the data decoding will be detected before the time of the decoding process. All macroblocks in the data packet in which a fully decoded macroblock is located are determined to be abnormal macroblocks.
  • the decoder can also typically handle a syntax error in the code stream.
  • the code stream abnormality determining module 20 is configured to: when the decoder performs a decoding process, a syntax error occurs when a syntax error occurs. All macroblocks in the data packet are determined to be abnormal macroblocks.
  • the code stream abnormality determining module 20 detects that the syntax error refers to one or more of the following situations:
  • the motion vector in the macro block exceeds the valid range, and the variable length code (Variable Length Code, referred to as VLC) Invalid, Discrete Cosine Transform (DCT) coefficient value exceeds the effective range, the number of discrete cosine transform coefficients exceeds the effective range, the quantized coefficient exceeds the valid range, and the number of decoded macroblocks exceeds the valid range.
  • VLC Variable Length Code
  • DCT Discrete Cosine Transform
  • the data packet referred to in the present invention refers to one frame of image data specified in the video compression standard, or a macroblock group, or a video pack, or a stripe.
  • the macroblock group typically refers to the Groups of Blocks (GOB) specified in the H.263 standard
  • the video packet typically refers to the video packet (Video Packet, VP for short) specified in the MPEG-4 standard.
  • the band is typically a strip defined in the 264.264 standard.
  • methods for improving the quality of video decoded images include:
  • Step 30 detecting a code stream abnormality during decoding of the video image compressed data stream
  • Step 32 Conceal the abnormal macroblock.
  • Typical case 1 When data loss is detected during decoding, data loss will be detected. All macroblocks in the data packet in which the last fully decoded macroblock is located in the decoding process before the moment is determined as an abnormal macroblock.
  • the macroblock between the S macroblock and the M macroblock belongs to the same data packet X.
  • the decoder After decoding the E macroblock and decoding the E macroblock completely, the decoder does not find a syntax error, and the decoding is continued when decoding is continued.
  • the N macroblock the decoder determines the macroblock loss between the E macroblock and the M macroblock, in order to prevent the code stream error from causing an error in the data decoded by the macroblock before the E macroblock, thereby causing the color block or the macroblock to be misaligned, etc. All the macroblocks in the data packet X in which the E macroblock is located are determined as abnormal macroblocks.
  • Typical case 2 When a syntax error is detected for the code stream during decoding, all macroblocks in the data packet in which the syntax error macroblock is located are determined as abnormal macroblocks.
  • the detection of a syntax error means that one or more of the following conditions are detected: The motion vector in the macroblock is out of the valid range, the variable length codeword is invalid, the value of the discrete cosine transform coefficient is outside the valid range, and the number of discrete cosine transform coefficients is more than valid.
  • the range, the quantized coefficient is out of the valid range, and the number of decoded macroblocks is out of the valid range.
  • the macroblock between the A macroblock and the E macroblock belongs to the same data packet Y, and a syntax error is detected when the decoder decodes the D macroblock, and the macroblock corresponding to the D macroblock is decoded correspondingly. Number It may be wrong, but the error mode does not belong to the error mode in the syntax detection algorithm in the decoder, so all the macroblocks in the data packet Y, which is the data block of the D macroblock, are determined as abnormal macroblocks.
  • the concealing process performed on the abnormal macroblock in the present invention is a prior art, and there are various ways of concealing processing, for example, replacing an abnormal macroblock with a determined fixed macroblock, and performing pixel value blurring processing on the abnormal macroblock. , Equalize the pixel values in the abnormal macroblock, and so on.
  • the data packet referred to in the present invention refers to one frame of image data specified in the video compression standard, or a macroblock group, or a video pack, or a stripe.
  • the macroblock group typically refers to the Groups of Blocks (GOB) specified in the H.263 standard
  • the video packet typically refers to the video packet (Video Packet, VP for short) specified in the MPEG-4 standard.
  • the band typically refers to the strip specified in the H.264 standard.
  • the determined abnormal macro block is in units of data packets, thereby effectively eliminating color blocks or misaligned isolated blocks appearing in video image decoding caused by packet loss or bit error. phenomenon.
  • the adverse effects of abnormal macroblocks with similar positions or similar positions on the decoding process of adjacent normal macroblocks and the decoding process of abnormal macroblocks for subsequent normal macroblocks are effectively suppressed. Influence, improve the overall viewing effect of the decoded video, and improve the user experience.
  • the determined abnormal macro block is in units of data packets, thereby effectively eliminating color patches or misalignments occurring in video image decoding caused by packet loss or bit error. Isolated blocks and other phenomena. Especially in the case of poor code stream quality, the adverse effects of abnormal macroblocks with similar positions or similar positions on the decoding process of adjacent normal macroblocks and the decoding process of abnormal macroblocks for subsequent normal macroblocks are effectively suppressed. Influence, improve the overall viewing effect of the decoded video, and improve the user experience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

本发明公开了一种提高视频解码图像质量的方法及解码器，此方法包括：对视频图像压缩数据流进行解码的过程中，检测到码流异常后确定异常宏块并且确定出的异常宏块以数据包为单位，并对异常宏块进行隐蔽处理。本发明的解码过程中检测到码流异常时，确定出的异常宏块以数据包为单位，从而有效的消除因丢包或误码引起的视频图像解码中出现的色块或错位孤立块等现象。尤其在码流质量较差的情况下，有效的抑制了位置相连或位置相近的异常宏块对相邻正常宏块的解码过程的不良影响以及异常宏块对后续正常宏块的解码过程的不良影响，提高解码视频的整体观看效果，提高用户体验。

Description

一种提高视频解码图像质量的方法及解码器
技术领域
本发明属于视频通讯领域， 具体涉及一种提高视频解码图像质量的方法 及解码器。
背景技术
在无线视频通讯中， 多径衰落现象或外部干扰信号容易使无线信号不稳 定， 传输质量下降， 并且使得视频码流中存在丟包和误码的现象。 在目前第 三代移动通信（ 3rd Generation , 简称 3G )技术标准使用的 H.324-M规范中 已规定使用 H.263、 H.264和 MPEG-4 ( Moving Picture Experts Group )视频编 码标准。 这几种标准都是基于运动补偿的视频编码算法， 其中的大多数视频 帧与前面帧的相关性很高， 任何一帧的错误都会被传播、 扩散到后续的若干 帧中， 从而引起视频质量的严重降低。
码流丟包直接导致了解码信息的丟失， 使得解码出现错误， 而码流误码 则会导致解码信息的错误， 同样会影响解码视频质量。 另外， 在码流丟包时， 如果解码器没有及时发现丟包的发生， 而误将后续的码流包作为当前码流进 行解码， 则会出现与码流误码类似的后果， 主要的现象就是解码时会出现严 重的色块、 错位孤立块等。
一般来说码流误码有多种不同的表现形式， 例如头错误、 运动矢量错误 和量化系数错误。 头错误会造成后面的数据都不能正常解码（包括运动矢量 和量化系数） 。 运动矢量错误会造成运动补偿位置不正确， 产生孤立块。 量 化系数错误会造成解码的亮度或色度值不正确。
实际解码过程中异常码流对应的异常宏块经常位置相连或位置相近， 位 置相近的异常宏块会影响与其相邻的正常宏块的解码过程， 并且异常宏块还 可能导致其后续正常宏块解码错误。 另外， 已有技术中在解码过程中判断异 常宏块时一般具有滞后性，如图 1所示，在码流中宏块 MBn已经是错误宏块， 但此宏块 MBn不符合解码器判断异常宏块的评判条件， 可能导致解码器在解 码到宏块 MBn+m时才发现错误， 从而使 MBn到 MBn+n>1之间的宏块解码后， 图像出现色块或错位孤立块的现象。
发明内容
本发明要解决的技术问题是提供一种提高视频解码图像质量的方法及解 码器， 有效消除因丟包或误码导致的解码图像中色块或错位孤立块的现象， 提高解码图像的质量。
为了解决上述技术问题， 本发明提供了一种提高视频解码图像质量的方 法， 包括： 对视频图像压缩数据流进行解码的过程中， 检测到码流异常后确 处理。
可选地， 上述方法还可以具有以下特点：
所述检测到码流异常后确定异常宏块的步骤包括： 解码时检测到数据丟 失情况时， 将检测到数据丟失这一时刻之前的解码过程中最后一个得到完整 解码的宏块所在的数据包中所有宏块确定为异常宏块。
可选地， 上述方法还可以具有以下特点：
所述检测到码流异常后确定异常宏块的步骤包括： 解码过程中针对码流 检测到语法错误时， 将出现语法错误的宏块所在的数据包中所有宏块确定 为异常宏块。
可选地， 上述方法还可以具有以下特点：
检测到语法错误是指检测到以下情况中一种或几中： 宏块中运动矢量超 出有效范围， 变长码字无效、 离散余弦变换系数的值超出有效范围， 离散余 弦变换系数个数超出有效范围、 量化系数超出有效范围、 解码出的宏块数目 超出有效范围。
可选地， 上述方法还可以具有以下特点：
所述数据包是指视频压缩标准中规定的一帧图像数据、 或一个宏块组、 或一个视频包、 或一个条带。 为了解决上述技术问题， 本发明还提供了一种提高视频解码图像质量的 解码器， 所述解码器包括码流异常判定模块和隐蔽处理模块； 所述码流异常 判定模块设置为： 在解码器对视频图像压缩数据流进行解码的过程中， 检测 常宏块的位置通知至所述隐蔽处理模块； 所述隐蔽处理模块设置为： 对从所 述码流异常判定模块获知的异常宏块位置处的数据进行隐蔽处理。
可选地， 上述解码器还可以具有以下特点：
所述码流异常判定模块是设置为以如下方式检测到码流异常后确定异常 宏块在所述解码器进行解码时检测到数据丟失情况时， 将检测到数据丟失这 一时刻之前的解码过程中最后一个得到完整解码的宏块所在的数据包中所有 宏块确定为异常宏块。
可选地， 上述解码器还可以具有以下特点：
所述码流异常判定模块是设置为以如下方式检测到码流异常后确定异常 宏块在所述解码器进行解码过程中针对码流检测到语法错误时， 将出现语 法错误的宏块所在的数据包中所有宏块确定为异常宏块。
可选地， 上述解码器还可以具有以下特点：
所述码流异常判定模块检测到语法错误是指检测到以下情况中一种或几 中： 宏块中运动矢量超出有效范围， 变长码字无效、 离散余弦变换系数的值 超出有效范围， 离散余弦变换系数个数超出有效范围、 量化系数超出有效范 围、 解码出的宏块数目超出有效范围。
可选地， 上述解码器还可以具有以下特点：
所述数据包是指视频压缩标准中规定的一帧图像数据、 或一个宏块组、 或一个视频包、 或一个条带。
本发明的解码过程中检测到码流异常时，确定出的异常宏块以数据包为 单位， 从而有效的消除因丟包或误码引起的视频图像解码中出现的色块或 错位孤立块等现象。 尤其在码流质量较差的情况下， 有效的抑制了位置相连 或位置相近的异常宏块对相邻正常宏块的解码过程的不良影响以及异常宏块 对后续正常宏块的解码过程的不良影响， 提高解码视频的整体观看效果， 提 高用户体验。 附图概述
图 1是已有技术中判断异常宏块时出现滞后情况的示意图；
图 2是本发明实施例中提高视频解码图像质量的解码器的组成结构图； 图 3是本发明实施例中提高视频解码图像质量的方法流程图；
图 4是本发明实施例中典型实施方式一的解码示例附图；
图 5是本发明实施例中典型实施方式二的解码示例附图。 本发明的较佳实施方式
如图 2所示， 提高视频解码图像质量的解码器包括码流异常判定模块 20 和隐蔽处理模块 21。
码流异常判定模块 20设置为：在解码器对视频图像压缩数据流进行解码 为单位， 并将异常宏块的位置通知至所述隐蔽处理模块。
隐蔽处理模块 21设置为：对从所述码流异常判定模块获知的异常宏块位 置处的数据进行隐蔽处理。
本解码器可典型处理处理数据丟失情况， 码流异常判定模块 20是设置 为： 在所述解码器进行解码时检测到数据丟失情况时， 将检测到数据丟失这 一时刻之前的解码过程中最后一个得到完整解码的宏块所在的数据包中所有 宏块确定为异常宏块。
本解码器还可典型处理码流出现语法错误的情况， 码流异常判定模块 20 是设置为：在所述解码器进行解码过程中针对码流检测到语法错误时，将出 现语法错误的宏块所在的数据包中所有宏块确定为异常宏块。
码流异常判定模块 20检测到语法错误是指检测到以下情况中一种或几 中： 宏块中运动矢量超出有效范围， 变长码字 （Variable Length Code, 简称 VLC )无效、 离散余弦变换 (Discrete Cosine Transform, 简称 DCT)系数的值超 出有效范围， 离散余弦变换系数个数超出有效范围、 量化系数超出有效范围、 解码出的宏块数目超出有效范围。
本发明中所指的数据包是指视频压缩标准中规定的一帧图像数据、 或一 个宏块组、 或一个视频包、 或一个条带。 其中， 宏块组典型是指 H.263标准 中规定的宏块组（Groups of Blocks, 简称 GOB ) , 视频包典型是指 MPEG-4 标准中规定的视频包（Video Packet, 简称 VP ) , 条带典型是指 Η.264标准 中规定条带 （Slice ) 。
如图 3所示， 提高视频解码图像质量的方法包括：
步骤 30: 对视频图像压缩数据流进行解码的过程中， 检测到码流异常；
步骤 32: 对异常宏块进行隐蔽处理。
典型适用的情况一： 解码时检测到数据丟失情况时， 将检测到数据丟失 这一时刻之前的解码过程中最后一个得到完整解码的宏块所在的数据包中所 有宏块确定为异常宏块。
如图 4所示， S宏块至 M宏块间的宏块属于同一数据包 X, 解码器解 码 E宏块并对 E宏块进行完整解码后未发现语法错误， 继续解码时解码出 的是 N宏块， 解码器判定 E宏块至 M宏块间的宏块丟失， 为了防止码流 错误导致 E宏块之前的宏块解码出的数据出现错误从而导致色块或宏块错 位等情况，将 E宏块所在的数据包即数据包 X中所有宏块确定为异常宏块。
典型适用的情况二： 解码过程中针对码流检测到语法错误时， 将出现 语法错误的宏块所在的数据包中所有宏块确定为异常宏块。检测到语法错误 是指检测到以下情况中一种或几中： 宏块中运动矢量超出有效范围， 变长码 字无效、 离散余弦变换系数的值超出有效范围， 离散余弦变换系数个数超出 有效范围、 量化系数超出有效范围、 解码出的宏块数目超出有效范围。
如图 5所示， A宏块至 E宏块间的宏块属于同一数据包 Y, 解码器解 码到 D宏块时检测到语法错误，此时 D宏块之前的宏块对应的解码出的数 据可能是错误的，但误码方式不属于解码器中语法检测算法中的误码方式， 所以将 D宏块所在的数据包即数据包 Y中所有宏块确定为异常宏块。
本发明中对异常宏块所进行的隐蔽处理是已有技术， 进行隐蔽处理的方 式有多种， 例如将异常宏块替换为已确定的固定宏块， 对异常宏块的进行像 素值模糊处理， 对异常宏块中像素值进行均衡处理等。
本发明中所指的数据包是指视频压缩标准中规定的一帧图像数据、 或一 个宏块组、 或一个视频包、 或一个条带。 其中， 宏块组典型是指 H.263标准 中规定的宏块组（Groups of Blocks, 简称 GOB ) , 视频包典型是指 MPEG-4 标准中规定的视频包（Video Packet, 简称 VP ) , 条带典型是指 H.264标准 中规定的条带 ( Slice ) 。
本发明的解码过程中检测到码流异常时，确定出的异常宏块以数据包为 单位， 从而有效的消除因丟包或误码引起的视频图像解码中出现的色块或 错位孤立块等现象。 尤其在码流质量较差的情况下， 有效的抑制了位置相连 或位置相近的异常宏块对相邻正常宏块的解码过程的不良影响以及异常宏块 对后续正常宏块的解码过程的不良影响， 提高解码视频的整体观看效果， 提 高用户体验。
当然， 本发明还可有其他多种实施例， 在不背离本发明精神及其实质的 但这些相应的改变和变形都应属于本发明所附的权利要求的保护范围。 本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序 来指令相关硬件完成， 所述程序可以存储于计算机可读存储介质中， 如只读 存储器、 磁盘或光盘等。 可选地， 上述实施例的全部或部分步骤也可以使用 一个或多个集成电路来实现。 相应地， 上述实施例中的各模块 /单元可以釆用 硬件的形式实现， 也可以釆用软件功能模块的形式实现。 本发明不限制于任 何特定形式的硬件和软件的结合。 工业实用性 本发明的解码过程中检测到码流异常时，确定出的异常宏块以数据包为 单位， 从而有效的消除因丟包或误码引起的视频图像解码中出现的色块或 错位孤立块等现象。 尤其在码流质量较差的情况下， 有效的抑制了位置相连 或位置相近的异常宏块对相邻正常宏块的解码过程的不良影响以及异常宏块 对后续正常宏块的解码过程的不良影响， 提高解码视频的整体观看效果， 提 高用户体验。

Claims

权 利 要 求 书
1、 一种提高视频解码图像质量的方法， 其包括：
对视频图像压缩数据流进行解码的过程中， 检测到码流异常后确定异常 宏块并且确定出的异常宏块以数据包为单位， 并对异常宏块进行隐蔽处理。
2、 如权利要求 1所述的方法， 其中，
所述检测到码流异常后确定异常宏块的步骤包括： 解码时检测到数据丟 失情况时， 将检测到数据丟失这一时刻之前的解码过程中最后一个得到完整 解码的宏块所在的数据包中所有宏块确定为异常宏块。
3、 如权利要求 1所述的方法， 其中，
所述检测到码流异常后确定异常宏块的步骤包括： 解码过程中针对码流 检测到语法错误时， 将出现语法错误的宏块所在的数据包中所有宏块确定 为异常宏块。
4、 如权利要求 3所述的方法， 其中，
检测到语法错误是指检测到以下情况中一种或几中： 宏块中运动矢量超 出有效范围， 变长码字无效、 离散余弦变换系数的值超出有效范围， 离散余 弦变换系数个数超出有效范围、 量化系数超出有效范围、 解码出的宏块数目 超出有效范围。
5、 如权利要求 1至 4中任一权利要求所述的方法， 其中，
所述数据包是指视频压缩标准中规定的一帧图像数据、 或一个宏块组、 或一个视频包、 或一个条带。
6、一种提高视频解码图像质量的解码器， 所述解码器包括码流异常判定 模块和隐蔽处理模块；
所述码流异常判定模块设置为： 在解码器对视频图像压缩数据流进行解 包为单位， 并将异常宏块的位置通知至所述隐蔽处理模块；
所述隐蔽处理模块设置为： 对从所述码流异常判定模块获知的异常宏块 位置处的数据进行隐蔽处理。
7、 如权利要求 6所述的解码器， 其中，
所述码流异常判定模块是设置为以如下方式检测到码流异常后确定异常 宏块： 在所述解码器进行解码时检测到数据丟失情况时， 将检测到数据丟失 这一时刻之前的解码过程中最后一个得到完整解码的宏块所在的数据包中所 有宏块确定为异常宏块。
8、 如权利要求 6所述的解码器， 其中，
所述码流异常判定模块是设置为以如下方式检测到码流异常后确定异常 宏块： 在所述解码器进行解码过程中针对码流检测到语法错误时， 将出现 语法错误的宏块所在的数据包中所有宏块确定为异常宏块。
9、 如权利要求 8所述的解码器， 其中，
所述码流异常判定模块检测到语法错误是指检测到以下情况中一种或几 种： 宏块中运动矢量超出有效范围， 变长码字无效、 离散余弦变换系数的值 超出有效范围， 离散余弦变换系数个数超出有效范围、 量化系数超出有效范 围、 解码出的宏块数目超出有效范围。
10、 如权利要求 6至 9中任一权利要求所述的解码器， 其中，
所述数据包是指视频压缩标准中规定的一帧图像数据、 或一个宏块组、 或一个视频包、 或一个条带。
PCT/CN2011/075582 2010-06-30 2011-06-10 一种提高视频解码图像质量的方法及解码器 WO2012000379A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201010222763.2A CN101888562B (zh) 2010-06-30 2010-06-30 一种提高视频解码图像质量的方法及解码器
CN201010222763.2 2010-06-30

Publications (1)

Publication Number Publication Date
WO2012000379A1 true WO2012000379A1 (zh) 2012-01-05

Family

ID=43074242

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/075582 WO2012000379A1 (zh) 2010-06-30 2011-06-10 一种提高视频解码图像质量的方法及解码器

Country Status (2)

Country Link
CN (1) CN101888562B (zh)
WO (1) WO2012000379A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101888562B (zh) * 2010-06-30 2014-03-19 中兴通讯股份有限公司 一种提高视频解码图像质量的方法及解码器
CN104202611A (zh) * 2014-09-01 2014-12-10 赛特斯信息科技股份有限公司 基于视频解码器实现视频文件中花屏缺陷检测的方法
CN109040761A (zh) * 2018-09-17 2018-12-18 俞群爱 一种带有加密水印的监控上墙系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1295411A (zh) * 1999-09-30 2001-05-16 松下电器产业株式会社 活动图象解码方法、活动图象解码装置及程序记录媒体
CN101188772A (zh) * 2006-11-17 2008-05-28 中兴通讯股份有限公司 一种视频解码的时域错误隐蔽方法
CN101304538A (zh) * 2008-06-19 2008-11-12 宁波中科集成电路设计中心有限公司 Avs视频解码的错误检测与掩盖方法
CN101888562A (zh) * 2010-06-30 2010-11-17 中兴通讯股份有限公司 一种提高视频解码图像质量的方法及解码器

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1295411A (zh) * 1999-09-30 2001-05-16 松下电器产业株式会社 活动图象解码方法、活动图象解码装置及程序记录媒体
CN101188772A (zh) * 2006-11-17 2008-05-28 中兴通讯股份有限公司 一种视频解码的时域错误隐蔽方法
CN101304538A (zh) * 2008-06-19 2008-11-12 宁波中科集成电路设计中心有限公司 Avs视频解码的错误检测与掩盖方法
CN101888562A (zh) * 2010-06-30 2010-11-17 中兴通讯股份有限公司 一种提高视频解码图像质量的方法及解码器

Also Published As

Publication number Publication date
CN101888562B (zh) 2014-03-19
CN101888562A (zh) 2010-11-17

Similar Documents

Publication Publication Date Title
JP5007012B2 (ja) ビデオ符号化方法
AU774040B2 (en) Video coding
US8743970B2 (en) Video decoding with error detection and concealment
US20110051812A1 (en) Video Transmitting Apparatus, Video Receiving Apparatus, Video Transmitting Method, and Video Receiving Method
WO2003007495A1 (en) A method and device for suspecting errors and recovering macroblock data in video coding
WO2003061284A1 (en) Motion wake identification and control mechanism
JP2009081576A (ja) 動画像復号装置、および動画像復号方法
WO2012000379A1 (zh) 一种提高视频解码图像质量的方法及解码器
JP2009081579A (ja) 動画像復号装置、および動画像復号方法
US20050089102A1 (en) Video processing
JP4432582B2 (ja) 動画像情報復元装置、動画像情報復元方法、動画像情報復元プログラム
Huang et al. A video data recovery algorithm in wireless communication networks
JP4570081B2 (ja) 動画像エラー隠蔽方法および装置
US20050123047A1 (en) Video processing
JP4309784B2 (ja) 復号器
JP2011239464A (ja) 復号器
JP2009105986A (ja) 復号器
KR100557118B1 (ko) 동영상 디코더 및 이를 이용한 디코딩 방법
JPH0937244A (ja) 動画像データ誤り検出装置
Superiori et al. Cross-layer detection of visual impairments in h. 264/avc video sequences streamed over umts networks
KR20050089457A (ko) 동영상 복호화 시 에러 은닉 방법
KR100590328B1 (ko) 동영상 디코더 및 이를 이용한 디코딩 방법
JP2003087792A (ja) 動画像符号化装置
KR20050099079A (ko) 동영상 복호화 시 에러 은닉 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11800124

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11800124

Country of ref document: EP

Kind code of ref document: A1