CN101094406B - Method and device for transferring video data stream - Google Patents

Method and device for transferring video data stream Download PDF

Info

Publication number
CN101094406B
CN101094406B CN 200710119403 CN200710119403A CN101094406B CN 101094406 B CN101094406 B CN 101094406B CN 200710119403 CN200710119403 CN 200710119403 CN 200710119403 A CN200710119403 A CN 200710119403A CN 101094406 B CN101094406 B CN 101094406B
Authority
CN
China
Prior art keywords
data stream
unit
rate
quantization
video data
Prior art date
Application number
CN 200710119403
Other languages
Chinese (zh)
Other versions
CN101094406A (en
Inventor
张�浩
Original Assignee
北京中星微电子有限公司
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 北京中星微电子有限公司 filed Critical 北京中星微电子有限公司
Priority to CN 200710119403 priority Critical patent/CN101094406B/en
Publication of CN101094406A publication Critical patent/CN101094406A/en
Application granted granted Critical
Publication of CN101094406B publication Critical patent/CN101094406B/en

Links

Abstract

The method comprises: after making low bit rate quantization for the video stream treated with discrete cosine transformation (DCT), the video stream is encoded; after encoding the video stream, sending out a first data stream ; according to the data stream treated with low bit rate quantization, reconstructing the image information, and according to the reconstructed image information and said original image information of the video stream, getting the residual image information; after making DCT for the residual image information, making high bit rate quantization for the residual image information, and then making encode for it; after encode, sending out the second data stream. The apparatus thereof comprises: a DCT unit, a low bit rate quantization unit, a high bit rate quantization unit, an encode unit, a reconstruction unit and a residual data obtaining unit.

Description

一种视频数据流的传输方法及装置 A method and apparatus for transmitting video data stream

技术领域 FIELD

[0001] 本发明涉及数据流的传输方法,尤其是指一种视频数据流的传输方法及装置。 [0001] The present invention relates to a method for transmitting data streams, and particularly to a method for transmitting video data stream and an apparatus. 背景技术 Background technique

[0002] 随着计算机技术的飞速发展以及网络技术的广泛普及,广播多媒体和点播多媒体的应用也越来越普遍,出现了视频点播、准视频点播、时移电视点播以及视频即时通信等各种服务。 [0002] With the widespread popularity of computer technology and the rapid development of network technology, multimedia broadcast and on-demand multimedia applications are becoming more common, there has been a video-on-demand, near video on demand, time-shifted television on demand and all kinds of video instant messaging, etc. service. 为了保证服务的实时性和连贯性,这些服务通常都需要比较大的网络带宽来传输大量的图像与视频数据。 In order to ensure real-time and continuity of care, these services often require large network bandwidth to transmit large amounts of image and video data. 因此,为了节约有限的带宽,人们通常使用编码技术将原始图像和视频数据在传输前进行有效压缩。 Accordingly, in order to save a limited bandwidth, it is generally used an original image and encoding techniques for efficiently compressing video data before transmission.

[0003] 在视频数据流的传输过程中,被传输的数据流可以被压缩成各种格式,最常见的为联合图像专家组(JPEG,Joint Photographic Experts Group)格式和运动图像专家组(MPEG, Moving Picture Experts Group)格式。 [0003] In the transmission of the video data stream, the data stream to be transmitted can be compressed in various formats, the most common is the Joint Photographic Experts Group (JPEG, Joint Photographic Experts Group) format and a Moving Picture Experts Group (the MPEG, Moving Picture Experts Group) format. 其中,JPEG 是国际标准化组织(ISO)和国际电报电话咨询委员会(CCITT)联合成立的一个专家组织,该组织所制定的国际上通用的适用于彩色和单色多灰度连续色调静态图像的数据压缩编码标准被称为JPEG标准。 Which, JPEG is the International Organization for Standardization (ISO) and the International Telegraph and Telephone Consultative Committee (CCITT) established a joint expert organizations, the international generic for color and monochrome multi-grayscale continuous-tone still image data developed by the organization JPEG compression coding standards are called standard. 利用JPEG标准进行压缩可以获得较高的压缩比,并保持较好的信噪比,从而大大节省图像存储空间,降低通信带宽,因此是一个适用范围很广的图像数据压缩标准,其应用并不局限于静止图像,也可应用于对视频数据流帧内数据的处理。 Using a standard JPEG compression can achieve high compression ratio, and maintain a good signal to noise ratio, which saves image memory space, reduce the communication bandwidth, and therefore is suitable for a wide range of image data compression standard, which does not apply limited to a still image, also be applied to processing of video data in the frame data stream. 以下将以JPEG标准为例,简单介绍一下对视频数据流的处理方法。 The following will be the JPEG standard, for example, a brief treatment of the video data stream.

[0004] 图1所示为现有技术中对视频数据流进行处理的流程图。 [0004] Figure 1 is a flowchart of the video data stream is processed in the prior art. 如图1所示,利用JPEG 标准处理视频数据流的具体步骤如下: 1, using the standard JPEG video stream processing steps are as follows:

[0005] 步骤101,输入所需传输的YUV(亮度和色度,也称之为YCrCb)数据流。 [0005] Step 101, YUV input required transmission (luminance and chrominance, also referred to as a YCrCb) data stream.

[0006] 所述YUV数据流为视频数据流中的一种,其中,YUV是欧洲电视系统采用的一种颜色编码方法,用于优化彩色视频信号的传输,其优点在于只需占用极少的带宽,且可解决彩色电视机与黑白电视机的兼容问题。 [0006] The YUV data stream is a video data stream, wherein, YUV color encoding is a European television system using the method for optimizing the transmission of color video signals, which occupies only very little advantage in that Bandwidth, and black and white TV and color TV resolve compatibility issues. YUV数据流中的“Y”表示亮度(Luminance或Luma)信号;而“U”和“V”表示色度(Chrominance或Chroma)信号,用于描述影像色彩及饱和度,指定像素的颜色。 YUV data stream "Y" represents luminance (Luminance or Luma) signal; and "U" and "V" represent chroma (Chrominance or by Chroma) signal, used to describe the image color and saturation, the color of the pixel specified.

[0007] 步骤102,对上述YUV数据流进行离散余弦变换(DCT,Discrete CosineTransform)。 [0007] Step 102, the YUV data stream of the above-described discrete cosine transform (DCT, Discrete CosineTransform).

[0008] 在JPEG标准中,可将视频数据流中的每一个数据帧都看成是一张独立的图像。 [0008] In the JPEG standard, each data frame may be the video data stream are regarded as an independent image. 而对上述YUV数据流进行DCT变换的目的是为了将YUV数据流中的每一帧图像从空间域转换到频域中,即将图像的像素点数据用不同的频率系数来进行表示,从而得知图像的像素点数据在频域中的频率分布情况。 The purpose of the above-described DCT transformation YUV data stream for each frame of the image in the YUV data stream is converted from a spatial domain into the frequency domain, i.e. pixel data of an image using different frequencies to represent coefficients, so that pixel data of an image in the frequency distribution of the frequency domain. 具体的DCT变换过程为:从YUV数据流中读入一帧图像, 将该帧图像作为原始图像并按从左到右,从上到下的顺序对其像素点进行扫描编码,将其分割成一系列具有8X8个像素点的子块后逐块进行处理。 Specific DCT transform process: reading an image from the YUV data stream, the press frame image as an original image from left to right and top to bottom scan encode the pixel, which is divided into a after the 8X8 sub-block having a series of pixel points of the processed block by block. 所述的处理为:读取上述子块中每个像素点的像素值并形成一个8X8的像素数据矩阵,然后对该矩阵使用DCT函数进行运算,输出一个8X8的频率系数矩阵。 The process is: reading the pixel value of each pixel in the pixel sub-block to form a data matrix of 8X8, and DCT function using the matrix operation, the output frequency of a 8X8 coefficient matrix. 该频率系数矩阵的左上角是低频分量的频率系数值, The upper left corner frequency coefficient matrix is ​​the frequency coefficient values ​​of low frequency components,

4右下角为高频分量的频率系数值。 4 is the lower right corner frequency of high-frequency component coefficient value. 由于原始图像子块中像素点的像素值的变化一般都比较平缓,从频域上看,经过DCT变换后的原始图像的频率系数主要集中在低频部分,因此上述的频率系数矩阵左上角的数值较大,而右下角的数值较小,趋近于零值。 Due to changes in the pixel values ​​of the original image pixel sub-blocks are typically relatively flat, seen from the frequency domain, the frequency coefficients of the original image through the DCT transform are mainly concentrated in low frequency, so the above value of the upper left corner frequency coefficient matrix larger, and the value of the lower right corner is small, close to zero value. 经过DCT变换后的图像数据的信息量和变换前的信息量相等,并未减少,所以DCT变换为无失真的变换。 After an equal amount of information of the information before the DCT transform and the transformed image data, it is not decreased, the DCT transform is lossless transform.

[0009] 步骤103,对DCT变换后得到的系数矩阵进行量化。 [0009] Step 103, after the DCT coefficient matrix obtained is quantized.

[0010] 由于人类的眼晴对图像数据低频部分的信息(例如亮度信息)比对高频部分的信息(例如色差信息)具有更高的敏感度,因此可以根据上述的人类视觉特性,尽可能地保持图像中的有用或重要的信息,而同时又尽量丢弃那些冗余的或不重要的信息,从而实现对图像数据的压缩。 [0010] Since the human eye information (e.g., luminance) than the information (e.g., color difference information) has higher sensitivity, high-frequency portion of the image data of the low frequency portion can be based on the human visual characteristics, as far as possible Useful image held or important information, while at the same time try to discard information that is redundant or unimportant to achieve compression of image data.

[0011] 为了达到压缩图像数据的目的,可使用量化表对上述经过DCT变换的频率系数矩阵进行量化。 [0011] For the purposes of the compressed image data, using the quantization table described above may be frequency-coefficient matrix of quantized DCT transform pair. 所述的量化表是根据人类视觉系统特性和压缩图像类型的特点预先定义并优化的量化系数矩阵;所述的量化,就是将上述频率系数矩阵中的各个元素除以上述量化系数矩阵中相应位置的元素,并舍去小数部分。 The quantization tables are predefined and optimized according to human visual system characteristics and the characteristics of the compressed image type matrix of quantized coefficients; said quantization, said frequency is the coefficient matrix by dividing each element of the quantization matrix coefficient corresponding to the position elements, and removing the decimal portion. 通过上述的量化处理,可在保证一定质量的前提下,将原始图像中一部分对视觉效果影响不大的频率信息(例如亮度信息)所对应的频率系数的值变成连续的相同的0值。 Quantization by the above process, the quality can be guaranteed under the premise, the value of the frequency coefficients little effect on the visual effect of a portion of the original image frequency information (e.g., luminance) corresponding to the same consecutive becomes zero. 此时,在后续的数据传输中就可不用再逐位传输连续的多个0值,而只需传输多个0值的“个数”信息,待解码时再按该信息恢复为多个0值。 In this case, the subsequent data transmission can no longer transmitted by a plurality of consecutive 0-bit value, but only the transmission information "the number of" a plurality of values ​​0, then the information is restored to be decoded plurality 0:00 value. 例如可将“00000”用“50”表示,而在解码时再根据“50”将数据恢复为“00000”。 It may be, for example, "00000" with "50" while the re-decoding in accordance with "50" to restore data "00000." 因此,通过上述的量化可大幅压缩上述的频率系数矩阵。 Thus, by the above quantization can significantly compressing the frequency coefficient matrix. 由上可知,上述的量化处理是不可逆的有损压缩,这也是造成JPEG压缩信息损失的主要原因。 From the foregoing, the above-mentioned quantization process is irreversible lossy compression, JPEG compression which is the main cause information loss.

[0012] 步骤104,对量化后得到的矩阵系数进行编码并输出。 [0012] Step 104, the quantized coefficient matrix obtained is encoded and output.

[0013] 为了便于传输、存储和解码器进行解码,需要将编码后的图像数据组装成一帧一帧的数据进行传输,这样的数据通常被称为编码数据流。 [0013] For ease of transport, storage and a decoder decodes the image data needs to be assembled into the encoded data of one frame of a transmission, such data is commonly referred encoded data stream.

[0014] 图2为现有技术中对视频数据流进行处理的装置的结构示意图。 [0014] FIG. 2 is a schematic structural diagram of a video data stream processing apparatus of the prior art. 如图2所示,现有技术中对视频数据流进行处理的装置一般包括DCT单元201、量化单元202、编码单元203。 As shown in FIG apparatus, the prior art processes for the video data stream 2 includes a DCT unit 201 generally, a quantization unit 202, an encoding unit 203. 其中,DCT单元201对接收到的YUV数据流进行DCT变换,并将经过DCT变换后的数据流输入量化单元202 ;量化单元202对接收到的数据流进行量化处理,并将量化处理后的数据流输入编码单元203 ;编码单元203对接收到的数据流进行编码,输出编码数据流。 Wherein, received DCT unit 201 DCT transform YUV data stream, and after the DCT quantization unit 202, an input data stream; quantization unit 202, the received data stream are quantized, and the quantized data processing encoding stream input unit 203; the coding unit 203 of the received data stream outputs the encoded data stream.

[0015] 虽然所需传输的YUV数据流经过上述处理后变成了可直接传输的编码数据流,但仍然需要较大的带宽来传输上述的编码数据流,而且上述的传输过程还需要持续一段较长的时间以保证所传输的图像质量,从而对网络带宽的可用性和可用的带宽提出了很高的要求,即为了得到尽可能高的图像质量,需要尽可能地利用网络的传输带宽传输尽可能多的视频数据流。 [0015] Although the required transmission YUV data stream after the above processing into the encoded data stream can be transmitted directly, but still requires a large bandwidth to transmit the coded data stream described above, and the above-described transmission process on for some a longer time to ensure the quality of the transmitted image, and thus the availability of bandwidth available network bandwidth requirements made high, i.e., in order to obtain the highest possible image quality, requires the use of transmission bandwidth of the transmission network to make possible probably more video data stream. 而在实际情况中,由于网络用户数、传输数据量的涨落、网络的处理能力以及其它的因素,网络的可用带宽在不断发生变化。 In the actual case, since the number of network users, the data transfer amount fluctuation, network processing power and other factors, the available bandwidth of the network is constantly changing. 在此情况下,如果发送过多的视频数据流, 使得所发送的视频数据流超过了网络的可用带宽,则会造成视频数据的丢失,出现丢帧的现象,从而严重影响接收端所接收的视频数据流的动态效果。 In this case, if too many transmitted video data stream so that the video data stream transmitted exceeds the available bandwidth of the network, will result in loss of video data, frame dropping occurs, thus seriously affecting the receiving end the received dynamic effects of the video data stream. 因此,在实际应用中,发送端在发送视频数据流时通常都尽量调整所发送的数据量,留出一定的带宽余量以避免出现丢帧的现象。 Thus, in practical applications, usually try to adjust the amount of data transmitted when the transmitting end transmits the video data stream, a certain bandwidth margin left to avoid dropped frames. 但在数据编码剧烈变化的时候,发送端即使使用上述的这种带宽控制方法,也很难对所发送的数据量做到有效控制,因此非常容易出现所发送的数据量超出带宽的现象, 从而造成丢帧、错帧,严重影响视频数据流的动态质量。 However, when data encoding dramatic changes, even if the transmission side using the above bandwidth control method of this, it is difficult to achieve an effective control of the amount of data transmitted, so it is the phenomenon of the amount of data transmitted exceeds bandwidth prone to resulting in dropped frames, frame wrong, seriously affecting the dynamic quality of the video data stream.

5发明内容 5 SUMMARY OF THE INVENTION

[0016] 有鉴于此,本发明的主要目的在于提供一种视频数据流的传输方法及装置,从而在数据编码流剧烈变化时也不会出现丢帧等严重影响动态效果的问题。 [0016] In view of this, the main object of the present invention is to provide a method and an apparatus for transmitting a video data stream, such problems seriously affect the dynamic effects of dropped frames, etc. does not occur when encoding data streams dramatic changes.

[0017] 为达到上述目的,本发明的技术方案是这样实现的: [0017] To achieve the above object, the technical solution of the present invention is implemented as follows:

[0018] 一种视频数据流的发送方法,其特征在于,该方法至少包括: [0018] A method of transmitting video data stream, characterized in that the method comprises at least:

[0019] A、对经过离散余弦变换后的视频数据流根据低码率量化表进行低码率量化后编码,并发送该编码后的第一数据流; [0019] A, after the flow of the low-rate quantization of low bit rate coding quantization table according to the video data after the discrete cosine transform, and transmits the first encoded data stream;

[0020] B、根据所述低码率量化后的视频数据流重构图像信息,根据所述重构的图像信息以及所述视频数据流的原始图像信息,求取残差图像信息; [0020] B, the video data stream reconstruction image information of the low-rate quantization, the reconstructed image according to original image information and the information of the video data stream, the residual image information is obtained;

[0021] C、对所述残差图像信息进行离散余弦变换,对离散余弦变换后的视频数据流根据高码率量化表进行高码率量化后编码,并发送该编码后的第二数据流。 [0021] C, the residual image information to discrete cosine transform, the video data stream after the discrete cosine transform according to the high-rate quantization encoding high rate quantization table, a second data stream and transmitting the encoded .

[0022] 一种视频数据流的接收方法,其特征在于,该方法至少包括: [0022] The method of receiving a video data stream, characterized in that the method comprises at least:

[0023] A1、对所接收到的编码后的第一数据流或第二数据流进行解码,得到解码后的第一数据流或解码后的第二数据流; [0023] A1, the first data stream after the received encoded second data stream, or decoding, the first data stream to obtain decoded data stream or a second decoding;

[0024] 其中,所述编码后的第一数据流为根据低码率量化表进行低码率量化后编码得到的视频数据流;所述编码后的第二数据流为根据所述低码率量化后的视频数据流重构图像信息,并根据所述重构的图像信息以及所述视频数据流的原始图像信息,求取残差图像信息,然后对所述残差图像信息进行离散余弦变换,再对离散余弦变换后的视频数据流根据高码率量化表进行高码率量化后编码得到的视频数据流; [0024] wherein the first encoded data stream after the low-rate quantization in accordance with quantization table obtained by encoding a low bit rate video data stream; and a second data stream is encoded according to the code rate of the low video data stream a reconstructed image information after the quantization, the reconstructed image according to original image information and the information of the video data stream, the residual image information is obtained, and the residual image information of a discrete cosine transform , video data and then the video data stream, a discrete cosine transform according to the high-rate quantization high rate stream obtained by coding quantization table;

[0025] B1、根据解码后的第一数据流合成视频数据流;或者,根据解码后的第一数据流和与第一数据流相对应的第二数据流合成视频数据流。 [0025] B1, according to the first decoded data stream synthesizing video data stream; Alternatively, composite video stream data of the first data stream according to a second decoded data stream and a second data stream corresponding.

[0026] 一种视频数据流的发送装置,该装置包括:离散余弦变换单元,低码率量化单元, 高码率量化单元,编码单元,重构单元和残差求取单元;所述离散余弦变换单元,用于接收视频数据流或所述残差求取单元发送的残差图像信息,将接收到的视频数据流进行离散余弦变换后,发送给所述低码率量化单元,将接收到的残差图像信息进行离散余弦变换后,发送给所述高码率量化单元;所述低码率量化单元,用于将接收到的经过离散余弦变换的数据流根据低码率量化表进行低码率量化后,分别发送给所述编码单元和所述重构单元;所述高码率量化单元,用于将接收到的经过离散余弦变换的数据流根据高码率量化表进行高码率量化后,发送给所述编码单元;所述编码单元,用于对接收的经过低码率量化或高码率量化后的数据流进行编码,发送相应的编 [0026] A transmitting apparatus of the video data stream, the apparatus comprising: a discrete cosine transform unit, the quantization unit low bit rate, high-rate quantization unit, an encoding unit, a reconstruction unit, and obtaining a residual unit; the discrete cosine after transformation unit for receiving a video data stream or the residual residual image information obtaining unit is transmitted, the received video data stream is a discrete cosine transform, sent to the low-rate quantization unit, the received residual image information after the discrete cosine transform, sent to the high-rate quantization unit; the low rate quantization unit for converting the received data stream through a discrete cosine transform low rate quantization table according to the low after quantization bit rate, are transmitted to the encoding unit and the reconstruction unit; the high-rate quantization means for data received through the discrete cosine transform according to the high flow rate of high rate quantization table after the quantization, is sent to the encoding unit; the encoding unit for the data streams through low rate or high rate quantizer quantizing the received encoded, encoding and transmitting the corresponding 数据流;所述重构单元,用于对接收到的经过低码率量化的数据流进行重构,将重构后的图像信息发送给残差求取单元;所述残差求取单元,用于接收视频数据流和所述重构单元发送的重构后的图像信息,根据视频数据流中的原始图像信息和所接收的重构后的图像信息求残差,将残差图像信息发送给所述离散余弦变换单元。 Data stream; the reconstruction unit, for low bit-rate quantized through the received data stream reconstruction, transmits the image information to the reconstructed residual obtaining unit; the residual obtaining unit, for receiving a video data stream and the reconstruction unit image information transmitted from the reconstruction request according to the residual image information of the original image information in the video data stream and the reconstruction of the received image information transmitting residual to the discrete cosine transform unit.

[0027] —种视频数据流的接收装置,其特征在于,该装置包括:解码单元和合成单元;所述解码单元,将接收到的编码后的第一数据流或第二数据流进行解码,得到解码后的第一数据流或解码后的第二数据流,发送给所述合成单元;其中,所述编码后的第一数据流为根据低码率量化表进行低码率量化后编码得到的视频数据流;所述编码后的第二数据流为根 [0027] - receiving means species of the video data stream, wherein, the apparatus comprising: a decoding unit and a synthesizing unit; said decoding unit, the first received data stream or the second coded data stream is decoded, the first data stream or decode the second data stream are decoded, sent to the synthesizing unit; wherein the first encoded data stream obtained after low-rate quantization the low rate coding quantization table video data stream; and a second data stream is encoded after the root

6据所述低码率量化后的视频数据流重构图像信息,并根据所述重构的图像信息以及所述视频数据流的原始图像信息,求取残差图像信息,然后对所述残差图像信息进行离散余弦变换,再对离散余弦变换后的视频数据流根据高码率量化表进行高码率量化后编码得到的视频数据流;所述合成单元,根据解码后的第一数据流合成视频数据流;或者,根据解码后的第一数据流和与第一数据流相对应的第二数据流合成为视频数据流,输出视频数据流。 6, according to the video data stream a reconstructed image information of low bit rate quantization, the reconstructed image according to original image information and the information of the video data stream, the residual image information is obtained, and then the residue difference discrete cosine transform image information, and then the video data stream discrete cosine transform video data stream obtained by encoding a high rate quantization the high rate quantization table; said synthesis unit, according to the first decoded data stream composite video data stream; or, according to a first data stream and the decoded second data stream with the synthesis of the first data stream corresponding to a video data stream, the output video data stream.

[0028] 综上可知,本发明提供了一种视频数据流的传输方法及装置,使得即使在数据编码流剧烈变化的时候也能保证基本图像信息的传输,从而不会造成丢帧等严重影响动态效果的问题;同时,在网络带宽及数据编码都正常的情况下也能够保证图像的精度和细腻质感。 [0028] To sum up, the present invention provides a method and apparatus for transmitting video data stream, such that even when the encoded data stream can be guaranteed dramatic changes in the basic information of the image transfer, so as not to severely affect the other frame loss problems dynamic effects; the same time, network bandwidth and data coding are normal, and it is possible to ensure the accuracy of delicate texture images.

[0029] 附图说明 [0029] BRIEF DESCRIPTION OF DRAWINGS

[0030] 图1为现有技术中对视频数据流进行处理的流程图。 [0030] FIG. 1 is a flowchart of the video data stream is processed in the prior art.

[0031] 图2为现有技术中对视频数据流进行处理的装置的结构示意图。 [0031] FIG. 2 is a schematic structural diagram of a video data stream processing apparatus of the prior art.

[0032] 图3为本发明实施例中发送视频数据流的流程图。 [0032] FIG. 3 is a flowchart of transmitting video data stream embodiment of the present invention.

[0033] 图4为本发明实施例中视频数据流发送装置的结构示意图。 [0033] FIG. 4 is a schematic structure of the video stream transmitting apparatus according to the present invention.

[0034] 图5为本发明实施例中视频数据流接收装置的结构示意图。 [0034] FIG. 5 is a schematic diagram of the structure of the video stream data receiving means of the embodiment of the present invention.

[0035] 具体实施方式 [0035] DETAILED DESCRIPTION

[0036] 为使本发明的目的、技术方案和优点表达得更加清楚明白,下面结合附图及具体实施例对本发明再作进一步详细的说明。 [0036] For purposes of the present invention, technical solutions and advantages clearer to obtain expression, in conjunction with the accompanying drawings and specific embodiments of the present disclosure is further described in detail.

[0037] 本发明提供了一种视频数据流的传输方法及装置,其主要思想是:在传输视频数据流之前,预先将所需传输的视频数据流中的每一个数据帧都分成基本层和高码率层;然后在视频数据流的传输过程中先利用小带宽完成当前帧的基本层的传输,实现当前帧的基本图像信息的传输;接着再利用剩余的带宽完成高码率层的传输,实现当前帧的图像高精度信息的传输,从而使得在上述视频数据流的传输过程中不会造成丢帧等严重影响动态效果的问题。 [0037] The present invention provides a method and apparatus for transmitting video data stream, the main idea is: before transmitting the video data stream, each data frame advance video data stream in the required transmission are divided into a base layer and high rate layer; then the transmission process in the video data stream to complete the transfer with a small bandwidth of the base layer of the current frame, to realize transmission of basic information of a current image frame; followed completed using the remaining bandwidth of high rate transmission layer , transmission of the image information with high accuracy to achieve the current frame, so that the transmission of the video data stream does not cause problems seriously affect dynamic effects frame dropping and the like.

[0038] 为了叙述的方便,在以下的实施例中,我们将以所传输的视频数据流为JPEG格式为例,对本发明实施例的技术方案进行详细地描述。 [0038] For convenience of description, in the following embodiments, we will be the transmitted video data stream format is JPEG as an example, the technical solutions of the embodiments of the present invention will be described in detail.

[0039] 图3为本发明实施例中发送视频数据流的流程图。 [0039] FIG. 3 is a flowchart of transmitting video data stream embodiment of the present invention. 在如下的处理过程中,我们将视频数据流(即YUV数据流)中的每一个数据帧都当成一张独立的图像进行处理。 In the following process, we each frame of a video data stream (i.e., YUV data stream) are treated as an independent image processing. 如图3 所示,本发明实施例中发送视频数据流的具体步骤如下: 3, the specific steps of transmitting video data stream embodiment of the present invention are as follows:

[0040] 步骤301,输入所需传输的YUV数据流。 [0040] Step 301, YUV data stream input desired transmission.

[0041 ] 步骤302,对上述YUV数据流进行DCT变换。 [0041] Step 302, the above-described DCT transformed YUV data stream. 即把YUV数据流中的每一个数据帧当成一张独立的图像,顺序地对视频数据流中的每一个数据帧进行DCT变换。 I.e., the YUV data for each data stream as a separate image frames sequentially to each data frame of the video data stream is a DCT transform. 为叙述方便,在以下所述的步骤303〜308中,以所处理的数据帧为同一个数据帧(可简称为当前帧)为例进行说明。 For convenience, in the following steps 303~308, the processed data frame to the same data frame as an example (abbreviated as the current frame) will be described.

[0042] 步骤303,对DCT变换后的数据帧进行低码率量化。 [0042] Step 303, after the data frame DCT transform low-rate quantization.

[0043] 所述的低码率量化为:设置一个量化表,并根据该量化表对经过上述DCT变换后的频率系数矩阵左上角的低频分量设置较细的量化(即使得频率系数的变化不大),而对其余部分即高频分量设置较粗的量化(即使得频率系数的变化较大),使得经过处理后的频率系数矩阵中的大部分频率系数很小,甚至为零,从而便于在以下的编码步骤中以较高的压缩比(例如30 : 1)对视频数据流中的数据帧进行压缩,形成该数据帧的基本层。 Low bit rate [0043] The quantization is: setting a quantization table, and the upper left corner of the matrix low-frequency components subjected to the above-described coefficients according to the quantization of DCT finer quantization table setting (frequency coefficients obtained even if no change large), i.e., the high frequency component set the quantization coarser (larger frequency coefficients obtained even if variation) of the rest, so that the frequency-coefficient matrix in the majority of the processed frequency coefficient is small, or even zero, so as to facilitate (e.g., 30: 1) in the following encoding step to a higher compression ratio of the video data stream in a data frame is compressed to form the base layer of the data frame. 上述的量化表可以是JPEG标准中标准的低码率量化表,也可以是根据具体情况自行设置的 The quantization table may be the above-described JPEG standard standard low rate quantization tables may be provided depending on the circumstances of their own

量化表。 Quantization table.

[0044] 步骤304,对低码率量化后的数据帧进行编码并输出基本层。 [0044] Step 304, the data frame after quantization of low bit rate and outputs coded base layer.

[0045] 所述的编码即为:对经过上述低码率量化后的数据帧进行压缩,使被压缩后的所需传输的数据量可以减到最小。 Encoding the [0045] namely: the data frame after quantization of the low bit rate compression, so that the amount of data required for transmission of the compressed can be minimized. 经过上述编码过程后,可得到当前帧的基本层,该基本层中携带了当前帧的基本图像信息。 After the encoding process, the base layer of the current frame is obtained, the base layer carries information of the current frame base image. 所述的编码可以分为变长编码、熵编码以及加入帧头信息等三个编码操作,所述的变长编码和熵编码为现有技术中常用的编码方式,所述的加入帧头信息包括利用当前帧帧头部分的保留字节,在所述保留字节中加入一个标志位,该标志位的作用为标志该基本层为当前帧的基本层。 The code can be divided into variable-length coding, and entropy coding header information added three coding operation, the variable-length coding and entropy coding is commonly used in the prior art encoding method, the frame header information is added using reserved bytes including a current frame header portion, a flag bit is added in the reserved byte, the effect of this flag is the flag of the base layer is a base layer of the current frame. 完成上述的编码操作后,根据具体的实际情况将上述基本层分割成一个或多个数据包,然后将这些数据包依次输出,形成该基本层的编码数据流,可以称之为第一数据流。 After completion of the encoding operation, depending on the above-described actual situation of the base layer is divided into one or more data packets, and sequentially outputs these packets to form a stream of the encoded base layer data, the first data stream may be referred to .

[0046] 步骤305,根据低码率量化后的数据帧重构基本层的图像信息。 [0046] Step 305, the data frame after quantization of low bit rate image information reconstructed base layer. 即根据步骤303中所述经过低码率量化后的数据帧,通过反量化以及反DCT变换等操作,重构上述基本层的图像信息。 I.e., step 303 through the low bit rate data frame according to quantization by the inverse quantization and inverse DCT operation and the like, the image information of the reconstructed base layer.

[0047] 步骤306,根据上述基本层的图像信息以及原始图像信息,求取残差。 [0047] Step 306, the image information of the base layer and the original image information, obtains the residual. 所述的原始图像信息即上述基本层相对应的数据帧的图像信息。 I.e., the original image information corresponding to the above-described base layer frame data of the image information. 所述的求残差,即用上述的原始图像信息减去上述的基本层的图像信息,得到相应的残差图像的信息。 The residual demand, i.e., subtracting the image information of the base layer with the original image information described above to give the corresponding residual image information. 该残差图像的信息中仍然存在数据冗余,因此还可用下述的步骤对残差图像的信息进行压缩。 The information is still present in the residual image data redundancy, and therefore the following step can also be used for the residual image information is compressed.

[0048] 步骤307,将上述残差图像进行DCT变换后,进行高码率量化。 After [0048] Step 307, the DCT transform the residual image, high-rate quantization.

[0049] 所述的DCT变换与步骤102和步骤302中的DCT变换相同。 [0049] The same DCT transform DCT transform step 102 and step 302. 所述的高码率量化为: 设置一个量化表,并根据该量化表对经过上述DCT变换后的频率系数矩阵的低频分量以及高频分量部分均设置较细的量化(即使得频率系数的变化不大),使得经过量化后的频率系数矩阵中的大部分频率系数都不为零,以便于在以下的编码步骤中以较低的压缩比(例如5 : 1)对残差图像进行压缩,形成该残差图像所属数据帧的高码率层。 The high rate quantization of: setting a quantization table, and the low frequency component and the high frequency component partial frequency coefficients after the DCT transform matrix are provided finer quantization (even if obtained based on the frequency variation coefficient quantization table little), so that after most of the frequency coefficient quantized frequency coefficient matrix is ​​not zero, so as to lower the compression ratio (e.g. 5 in the following encoding steps: 1 compression residual image), forming layer of the residual high bit rate image data frame belongs. 上述的量化表可以是JPEG标准中标准的高码率量化表,也可以是根据具体情况自行设置的量化表。 Quantization table described above may be a standard JPEG standard quantization table of high bit rate, the quantization table may be set depending on the circumstances of their own.

[0050] 步骤308,对高码率量化后的数据帧进行编码并输出。 [0050] Step 308, the data frame after quantization high rate encoded and output.

[0051] 所述的编码与步骤304中所述的编码方法类似,也可分为变长编码、熵编码以及加入帧头信息等三个部分。 [0051] The code is similar to step 304 of the coding method, it can be divided into three portions of variable length coding, entropy coding, and addition of header information. 经过上述编码过程后,可得到当前帧的高码率层,该高码率层中携带了当前帧的高精度的图像信息。 After the encoding process, the high rate of the current frame obtained layer, the layer carrying the high rate of precision of the current frame of the image information. 其中,进行变长编码和熵编码的方法与步骤304中变长编码和熵编码的方法相同;所述的加入帧头信息包括利用当前帧帧头部分的保留字节, 在所述保留字节中加入一个标志位,该标志位的作用为标志该高码率层为当前帧的高码率层。 Wherein, in the same manner as 304 and variable length coding and entropy coding method and variable length coding step of entropy coding; said added header information comprises reserved bytes using the current frame header portion, the reserved bytes adding a flag, the flag of the role flag layer is a high rate of the high rate level of the current frame. 完成上述的编码操作后,根据具体的实际情况将上述高码率层分割成一个或多个数据包,然后将这些数据包依次输出,形成该高码率层的编码数据流,可以称之为第二数据流。 After completion of the encoding operation, depending on the actual situation of the high bit rate layer is divided into one or more data packets, and sequentially outputs these packets forming the high rate encoded data stream layer, it may be referred to second data stream.

[0052] 在上述的步骤304和步骤308中,将所需传输的数据流的当前帧分为上述的基本层和高码率层后,可先利用小带宽完成当前帧的基本层的传输,然后再利用剩余的带宽完成当前帧的高码率层的传输。 [0052] In the above step 304 and step 308, the required data transmission after the current frame into the above-described base layer and a layer of high bit rate, the bandwidth available to complete the transfer with a small current frame of the base layer stream, and then use the remaining bandwidth of high rate transmission to complete the current layer frame. 在实际应用中,上述的同一个数据帧的基本层和高码率层是连续发送的,即先发送当前帧的基本层,接着发送当前帧的高码率层;然后再顺序发送下一个数据帧的基本层和高码率层,并依此类推直至完成所有数据流的传输。 In practical applications, the above-described data frame with a base layer and a high bit rate transmission layer is continuous, i.e., to transmit the base layer of the current frame, and then transmits the high bit rate layer of the current frame; and transmitting the sequence of data then the base layer and high layer frame rate, and so on until all the transmission data streams.

8[0053] 接收端在接收到包含基本层或高码率层的编码数据流后,可根据所接收的编码数据流中帧头信息的标志位得知所接收的编码数据流是基本层的编码数据流还是高码率层的编码数据流。 8 [0053] The receiving terminal comprises receiving the encoded data of the base layer after layer, or high rate flow, that may be encoded in the received data stream encoded according to the flag bit in the received data stream is a frame header information of the base layer encoded data stream or an encoded data stream layer high rate. 如果接收端所接收的编码数据流中包含基本层,则接收端可通过相应的解码装置对该编码数据流进行熵解码后再进行变长解码后得到该基本层所属数据帧的基本图像信息;同理,接收端也可以通过相应的解码装置对所接收到的高码率层的编码数据流进行相应解码后得到该高码率层所属数据帧的高精度的图像信息。 If the receiver received encoded data stream contains a base layer, the receiver may entropy decode the base image and then obtain information of the base layer data frame belongs to the variable-length decodes the coded data stream through the corresponding decoding means; after Similarly, the receiver can accordingly be performed to decode the received encoded data stream by a high rate layer corresponding decoding means to obtain highly accurate image data frame of the high rate information layer belongs. 如果接收端还能正确接收到该基本层所属数据帧的高码率层的信息,则可将所接收到的属于同一个数据帧的基本层和高码率层合并,从而获得该数据帧的完整图像信息;如果接收端由于网络带宽或其他的原因无法正确接收到某个数据帧的高码率层,则因为接收端已经接收到了该数据帧的基本层,所以能保证接收端能够得到该基本层所属数据帧的基本图像信息,因此虽然会影响到接收端所接收的图像信息的精度,但却不会产生丢帧的现象,从而不会对视频数据流的动态效果产生严重的影响。 If the receiver can correctly receive the information layer of the high rate data frame belongs to the base layer, it can be received belonging to the base layer frame is a high rate data and the layers were the same, thereby obtaining the data frame full image information; If the receiving terminal from the network bandwidth or other reason can not receive properly the layer of a high bit rate data frame, since the receiving end has received the base layer of the data frame, it is possible to ensure that the receiving end can be obtained basic image information of the base layer data frame belongs, so although it will affect the accuracy of the image information received by the receiver, but no frame dropping, so as not to have a serious impact on the dynamic effect of the video data stream.

[0054] 本发明实施例还提供了一种视频数据流发送装置。 [0054] Embodiments of the present invention further provides a video data stream transmitting apparatus. 图4为本发明实施例中视频数据流发送装置的结构示意图。 FIG 4 is a schematic structure of the video stream transmitting apparatus according to the present invention. 如图4所示,本发明实施例中的视频数据流发送装置包括: DCT单元401、低码率量化单元402、高码率量化单元403、编码单元404、重构单元405和残差求取单元406 ;所述编码单元404包括变长编码单元407、熵编码单元408、帧头信息加入单元409。 DCT unit 401, the quantization unit 402, the low bit rate, high-rate quantization unit 403, an encoding unit 404, and reconstruction unit 405 obtains a residual: 4, the present invention is a video data stream comprising transmitting apparatus in Embodiment unit 406; the coding unit 404 comprises a variable-length coding unit 407, an entropy coding unit 408, the header information unit 409 is added. 其中,所述的DCT单元401接收外界输入的YUV数据流或残差求取单元406发送的残差图像信息,将所接收的YUV数据流进行DCT变换后发送给所述低码率量化单元402, 将所接收的残差图像信息进行DCT变换后发送给所述高码率量化单元403 ;低码率量化单元402对所接收到的DCT变换后的数据流进行低码率量化,然后将低码率量化后的数据流分别发送给变长编码单元407和重构单元405 ;重构单元405对所接收到的低码率量化后的数据流进行重构,从而得到基本层的图像信息,并将基本层的图像信息发送给残差求取单元406 ;残差求取单元406根据所接收的基本层的图像信息以及该基本层所属的YUV数据流中的数据帧的原始图像信息求残差,并将残差信息发送给DCT单元401 ;所述高码率量化单元403对所接收到的DCT变换后的数据流进行高码率量化,然后将高码率量化后的数据流发 After which, a DCT unit 401 receives the YUV data stream or the residual external input unit obtains information transmitted residual image 406, the YUV data stream received DCT transform to a quantization unit 402, the low bit rate after the residual information of the received image DCT transform to a quantization unit 403, the high rate; low bit rate data stream quantization unit 402 of the received quantized DCT transform low bit rate, and low quantization bit rate data streams are transmitted to the variable length coding unit 407 and a reconstruction unit 405; the reconstructed data stream unit 405 to the received low rate quantization reconstruction to obtain image information of the base layer, and transmits the image information to the base layer residual obtaining unit 406; a residual demand obtaining unit 406 residues from the original image data in the YUV data stream of the received image information of the base layer and the base layer frame belongs information difference, and transmits the information to the residual DCT unit 401; a high rate data stream quantization unit 403 to the received high-rate quantization DCT transform, and the high rate data stream sent quantization 给变长编码单元407 ;变长编码单元407对所接收的量化后的数据流进行变长编码,然后将变长编码后的数据流发送给熵编码单元408 ;熵编码单元408对所接收的变长编码后的数据流进行熵编码,然后将熵编码后的数据流发送给帧头信息加入单元409 ;帧头信息加入单元409对所接收的熵编码后的数据流进行封装,并将相应的标志位写入帧头信息中,完成数据压缩,获得当前帧的基本层或高码率层并输出相应的编码数据流。 To the variable length coding unit 407; a data stream of variable-length coding unit 407 the received quantized variable-length coding, and the variable length coded data stream to the entropy encoding unit 408; entropy encoding unit 408 on the received data stream variable length coding entropy coding, and then transmits the data stream to the entropy coding unit 409 header information is added; the frame header information added to the data stream after the entropy encoding unit 409 encapsulates the received, and the corresponding the header information flag is written, data compression is completed, the high rate to obtain a base layer or layers of the current frame and outputs corresponding coded data stream.

[0055] 本发明实施例还提供了一种视频数据流接收装置。 Example [0055] The present invention also provides a video data stream receiving means. 图5为本发明实施例中视频数据流接收装置的结构示意图。 FIG 5 is a schematic diagram of the structure of the video stream data receiving means of the embodiment of the present invention. 如图5所示,本发明实施例中的视频数据流接收装置包括:解码单元501和合成单元502。 Synthesis unit 501 and decoding unit 502: 5, the present invention is a video data stream comprising receiving apparatus in embodiment.

[0056] 其中,所述解码单元501还包括帧头信息读取单元503、熵解码单元504和变长解码单元505。 [0056] wherein, the decoding unit 501 further includes a header information reading unit 503, entropy decoding unit 504 and the variable length decoding unit 505. 帧头信息读取单元503从所接收到的编码后的数据流中读取帧头信息中的标志位,将标志位信息发送给合成单元502,将所接收到的编码后的数据流发送给熵解码单元504 ;熵解码单元504对接收到的编码后的数据流进行熵解码,然后将熵解码后的数据流发送给变长解码单元505 ;变长解码单元505对接收到的熵解码后的数据流进行变长解码, 然后将变长解码后的数据流发送给合成单元502。 Header information reading unit 503 reads the flag in the frame header information from the data stream received encoded, the flag information sent to the synthesis unit 502, transmits the data stream to the received encoded entropy decoding unit 504; a data stream encoded entropy decoding unit 504 entropy decodes the received, and then transmits the data stream to the entropy decoding unit 505 variable-length decoding; after entropy decoding unit 505 variable-length decoding received the variable-length decoding the data stream, and then sends the data stream variable-length decoding to the combining unit 502. 所述合成单元502可根据所接收到的标 The synthesis unit 502 according to the received standard

9志位信息,得知所接收到的解码后的数据流为基本层还是高码率层。 9 Chi-bit information, that data stream to decode the received base layer or layer of high bit rate. 如果合成单元502所接收的解码后的数据流中包含同一个数据帧的基本层和高码率层,则合成单元502将上述同一个数据帧的基本层和高码率层进行合成,得到解码后的数据帧,该解码后的数据帧中包含了原始数据帧中的完整图像信息;如果合成单元502只接收到了某一个数据帧的基本层,而没有接收到该数据帧的高码率层,则合成单元502将上述的基本层合成为该数据帧的解码后的数据帧,该解码后的数据帧中包含了原始数据帧中的基本图像信息,因此虽然该解码后的数据帧并不包含原始数据帧中高精度的图像信息,对视频数据流的图像质量有一定的影响,但却避免了丢帧现象的出现,从而不会对视频数据流的动态效果产生严重的影响。 If the data stream synthesizing unit 502 decode the received data contains the same base layer frame rate and a high level, the synthesis unit 502 with a base layer of the data frame and the high bit rate layer synthesized by the decoding after the data frame, the data frame after the decoded contains complete image information of the original data frames; if the combining unit 502 receives the base layer one data frame, but does not receive the data frame the high bit rate layer , the data synthesis unit 502 frames the data for the base layer frame is synthesized decoded, the decoded data frame contains basic information of the original image data in the frame, so that while the data frame is not decoded frame containing image information in the original data with high accuracy, there is the image quality of the video data stream certain effect, but to avoid dropped frames so as not to have a serious impact on the dynamic effect of the video data stream.

[0057] 综上所述,本发明提供了一种视频数据流的传输方法及装置,使得在视频数据流的传输过程中,在数据码流以及网络带宽正常的时候能够保证接收端所接收到的视频数据流具有很高的精度和细腻质感,且在数据编码流剧烈变化的时候也能保证视频数据流中基本图像信息的传输,因此只会影响到视频数据流中图像的精度而不会出现丢帧等严重影响动态效果的现象。 [0057] In summary, the present invention provides a method and apparatus for transmitting video data streams, such that the transmission of the video data stream, the data stream in the normal time and network bandwidth can be received by the receiving end to ensure video data stream having a high accuracy and delicate texture, and the encoded data stream when drastic changes can also ensuring the transmission of video stream information of the base image, and therefore affects the precision of the video image data stream without phenomenon seriously affect the dynamic effects of dropped frames, etc. appear.

[0058] 以上所述,仅为本发明的较佳实施例而已,并非用于限定本发明的保护范围。 [0058] The above descriptions are merely preferred embodiments of the present invention but are not intended to limit the scope of the present invention. 凡在本发明的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。 Any modification within the spirit and principle of the present invention, made, equivalent substitutions, improvements, etc., should be included within the scope of the present invention.

10 10

Claims (11)

  1. 一种视频数据流的发送方法,其特征在于,该方法至少包括以下步骤:A、对经过离散余弦变换后的视频数据流根据低码率量化表进行低码率量化后编码,并发送该编码后的第一数据流;B、根据所述低码率量化后的视频数据流重构图像信息,根据所述重构的图像信息以及所述视频数据流的原始图像信息,求取残差图像信息;C、对所述残差图像信息进行离散余弦变换,对离散余弦变换后的视频数据流根据高码率量化表进行高码率量化后编码,并发送该编码后的第二数据流。 A method for transmitting video data stream, characterized in that the method comprises at least the following steps: A, the video data stream through the discrete cosine transform for the low bit-rate coding quantization of low bit rate in accordance with quantization table, and transmits the encoded a first data stream; B, the video data stream reconstruction image information of the low-rate quantization, the reconstructed image according to original image information and the information of the video data stream, the residual image is obtained information; C, the residual image information to discrete cosine transform, the video data stream a discrete cosine transform for the second data stream after the high rate quantization encoding and transmitting the high rate encoded according to a quantization table.
  2. 2.根据权利要求1所述的方法,其特征在于,步骤A和步骤C所述编码包括:在被编码的量化后的视频数据流中加入帧头信息。 2. The method according to claim 1, wherein said steps A and C coding comprises: frame header information added to the video data stream is encoded quantized.
  3. 3.根据权利要求2所述的方法,其特征在于,所述加入帧头信息包括:在所述第一数据流中的数据帧的帧头中加入一个标识该数据流为第一数据流的第一标志位;在所述第二数据流中的数据帧的帧头中加入一个标识该数据流为第二数据流的第二标志位。 3. The method according to claim 2, wherein said added header information includes: data in the first data stream to a frame header is added that identifies the data stream is a first data stream a first flag bit; the data in the second data stream to a frame header is added that identifies the data stream to a second data stream of the second flag bit.
  4. 4.根据权利要求3所述的方法,其特征在于,所述加入标志位包括:在数据帧的帧头部分的保留字节中加入第一标志位或第二标志位。 4. The method according to claim 3, wherein said flag is added comprising: adding a first flag or second flag bit in the reserved bytes header portion of a data frame.
  5. 5. 一种视频数据流的接收方法,其特征在于,该方法至少包括以下步骤:Al、对所接收到的编码后的第一数据流或第二数据流进行解码,得到解码后的第一数据流或解码后的第二数据流;其中,所述编码后的第一数据流为根据低码率量化表进行低码率量化后编码得到的视频数据流;所述编码后的第二数据流为根据所述低码率量化后的视频数据流重构图像信息,并根据所述重构的图像信息以及所述视频数据流的原始图像信息,求取残差图像信息, 然后对所述残差图像信息进行离散余弦变换,再对离散余弦变换后的视频数据流根据高码率量化表进行高码率量化后编码得到的视频数据流;Bi、根据解码后的第一数据流合成视频数据流;或者,根据解码后的第一数据流和与第一数据流相对应的第二数据流合成视频数据流。 First Al, the first data stream after the received encoded second data stream, or decoding, the decoded obtained: A method for receiving a video data stream, characterized in that the method comprises at least the steps of the second data stream, or decoding a data stream; wherein the encoded first data stream is a video data stream according to the low-rate quantization of low bit rate coding quantization table obtained; the encoded data of the second stream is a video data stream a reconstructed image according to the information of the low quantization bit rate, and based on the reconstructed image information and the image information of the original video data stream, the residual image information is obtained, then the residual discrete cosine transform image information, and then the video data stream discrete cosine transform video data stream obtained by encoding a high rate quantization the high rate quantization table; of Bi, composite video stream according to the first decoded data data stream; Alternatively, composite video stream data of the first data stream according to a second decoded data stream and a second data stream corresponding.
  6. 6.根据权利要求5所述的方法,其特征在于,步骤Al所述解码包括:从接收到的编码后的第一数据流中读取帧头信息中的第一标志位,从接收到的编码后的第二数据流中读取帧头信息中的第二标志位,对接收到的编码后的第一数据流或第二数据流进行解码。 6. The method as claimed in claim 5, wherein said step of decoding comprises Al: reading a first flag bit in the frame header information from the first data stream received coding, from the received the second data stream in the encoded second flag bit read header information, encoded first data stream or the second received data stream is decoded.
  7. 7.根据权利要求6所述的方法,其特征在于,所述步骤Bl中根据解码后的第一数据流和与第一数据流相对应的第二数据流合成视频数据流包括:B11、根据所述的第一标志位和第二标志位,将属于同一个数据帧的解码后的第一数据流和第二数据流合并,合成视频数据流。 7. The method according to claim 6, wherein said step Bl second data according to the first decoded data stream and a second data stream corresponding to the stream synthesizing video data stream comprising: B11, according to the first flag and second flag bits, the data stream belonging to the first and second data streams are combined with the decoded frame data, the composite video data stream.
  8. 8. 一种视频数据流的发送装置,其特征在于,该装置包括:离散余弦变换单元,低码率量化单元,高码率量化单元,编码单元,重构单元和残差求取单元;所述离散余弦变换单元,用于接收视频数据流或所述残差求取单元发送的残差图像信息,将接收到的视频数据流进行离散余弦变换后,发送给所述低码率量化单元,将接收到的残差图像信息进行离散余弦变换后,发送给所述高码率量化单元;所述低码率量化单元,用于将接收到的经过离散余弦变换的数据流根据低码率量化表进行低码率量化后,分别发送给所述编码单元和所述重构单元;所述高码率量化单元,用于将接收到的经过离散余弦变换的数据流根据高码率量化表进行高码率量化后,发送给所述编码单元;所述编码单元,用于对接收的经过低码率量化或高码率量化后的数据流进行编码,发送 An apparatus for transmitting video data stream, wherein, the apparatus comprising: a discrete cosine transform unit, the quantization unit low bit rate, high-rate quantization unit, an encoding unit, a reconstruction unit, and obtaining a residual unit; the described later, a discrete cosine transform unit for receiving a video data stream or the residual residual image information obtaining unit is transmitted, the received video data stream is a discrete cosine transform, sent to the low-rate quantization means, after the received residual discrete cosine transform image information, is sent to the high-rate quantization unit; the low rate quantization unit for receiving the flow of data through the discrete cosine transform according to the quantization of low bit rate after the low-rate quantization table, are transmitted to the encoding unit and the reconstruction means; the high-rate quantization means for data received through the discrete cosine transform according to the high flow rate quantization table after the high-rate quantization, is sent to the encoding unit; the encoding unit for the data streams through low rate or high rate quantizer quantizing the received encoded, transmitted 应的编码数据流;所述重构单元,用于对接收到的经过低码率量化的数据流进行重构,将重构后的图像信息发送给残差求取单元;所述残差求取单元,用于接收视频数据流和所述重构单元发送的重构后的图像信息, 根据视频数据流中的原始图像信息和所接收的重构后的图像信息求残差,将残差图像信息发送给所述离散余弦变换单元。 Corresponding coded data stream; the reconstruction unit, for low bit-rate quantized through the received data stream reconstruction, transmits the image information to the reconstructed residual obtaining unit; the residual seek fetch unit, for receiving a video data stream and the reconstruction unit image information transmitted from the reconstruction request according to the residual image information of the original image information in the video data stream received and reconstructed, the residual the image information sent to the discrete cosine transform unit.
  9. 9.根据根据权利要求8所述的装置,其特征在于,所述编码单元至少包括:变长编码单元,熵编码单元和帧头信息加入单元;所述变长编码单元,用于对接收到的经过低码率量化或高码率量化后的数据流变长编码,将变长编码后的数据流发送给所述熵编码单元;所述熵编码单元,用于对接收到的变长编码后的数据流进行熵编码,将熵编码后的数据流发送给所述帧头信息加入单元;所述帧头信息加入单元,用于在接收到的熵编码后的数据流的帧头中设置标志位,发送编码后的数据流。 9. The apparatus according to claim 8, wherein said encoding unit comprises at least: variable-length coding means, and entropy encoding unit cell header information is added; the variable length encoding unit for received rheological data length coding after quantization of low bit-rate quantized or high rate, the data stream transmitted to the variable length encoding unit entropy encoding; the entropy coding unit, for variable length encoding the received after entropy coding data stream, sends the data stream to the entropy encoding unit frame header information is added; the unit frame header information is added, after receiving the data for the entropy coding header stream provided flag, sending the encoded data stream.
  10. 10. 一种视频数据流的接收装置,其特征在于,该装置包括:解码单元和合成单元;所述解码单元,将接收到的编码后的第一数据流或第二数据流进行解码,得到解码后的第一数据流或解码后的第二数据流,发送给所述合成单元;其中,所述编码后的第一数据流为根据低码率量化表进行低码率量化后编码得到的视频数据流;所述编码后的第二数据流为根据所述低码率量化后的视频数据流重构图像信息,并根据所述重构的图像信息以及所述视频数据流的原始图像信息,求取残差图像信息, 然后对所述残差图像信息进行离散余弦变换,再对离散余弦变换后的视频数据流根据高码率量化表进行高码率量化后编码得到的视频数据流;所述合成单元,根据解码后的第一数据流合成视频数据流;或者,根据解码后的第一数据流和与第一数据流相对应的第二数据 A video apparatus receiving a data stream, wherein, the apparatus comprising: a decoding unit and a synthesizing unit; said decoding unit, the first data stream received encoded second data stream, or decoding, to obtain decoding the first data stream or second data stream after decoding, is sent to the synthesizing unit; wherein the first encoded data stream after the low-rate quantization the low rate coding quantization table obtained video data stream; and a second data stream according to the encoded video data stream a reconstructed image information after the quantization of low bit rate, and information of the original image and the reconstructed image information of the video data stream , the residual image information is obtained, then the residual discrete cosine transform image information, and then the video data stream, a discrete cosine transform high rate video data stream obtained by encoding the quantized quantization table according to the high bit rate; the synthesizing unit according to the first decoded data stream synthesizing video data stream; or second data according to the decoded first data stream and a second data stream corresponding to 合成为视频数据流,输出视频数据流。 Synthesis of a video data stream, the output video data stream.
  11. 11.根据权利要求10所述的装置,其特征在于,所述解码单元至少还包括:帧头信息读取单元、熵解码单元和变长解码单元;所述帧头信息读取单元,用于从接收到的编码后的第一数据流或第二数据流中读取帧头信息中的标志位,将标志位信息发送给所述合成单元,将接收到的编码后的第一数据流或第二数据流发送给所述熵解码单元;所述熵解码单元,用于对接收到的编码后的第一数据流或第二数据流熵解码,将熵解码后的第一数据流或第二数据流发送给所述变长解码单元;所述变长解码单元,用于对接收到的熵解码后的第一数据流或第二数据流变长解码, 将变长解码后的第一数据流或第二数据流发送给所述合成单元;所述合成单元,用于根据接收到的标志位信息和变长解码后的第一数据流合成视频数据流;或者,根据变长解码后的第一数据流和与 11. The apparatus according to claim 10, wherein at least said decode unit further comprises: header information reading unit, entropy decoding unit and the variable length decoding unit; the header information reading unit for flag read from the first frame header information in the received data stream or the second coded data stream, the flag information transmitted to the synthesizing unit, or the first data stream received encoded the second data stream to the entropy decoding unit; said entropy decoding unit, a first data stream for encoding the received data stream or the second entropy decoding of the first data stream or the entropy decoding after the first data stream after the first variable length decoding means for entropy decoding the received data or the second rheology-length decoding, the variable length decoding; transmitting said second data stream to variable-length decoding unit data stream or second data stream transmitted to the synthesizing unit; said synthesizing means for synthesizing the video data stream of the first data stream according to the received information and the flag variable length decoding; Alternatively, the variable length decoding according to the and a first data stream 一数据流相对应的第二数据流,合成视频数据流。 A data stream corresponding to the second data stream, the composite video data stream.
CN 200710119403 2007-07-23 2007-07-23 Method and device for transferring video data stream CN101094406B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200710119403 CN101094406B (en) 2007-07-23 2007-07-23 Method and device for transferring video data stream

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200710119403 CN101094406B (en) 2007-07-23 2007-07-23 Method and device for transferring video data stream

Publications (2)

Publication Number Publication Date
CN101094406A CN101094406A (en) 2007-12-26
CN101094406B true CN101094406B (en) 2010-09-29

Family

ID=38992379

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200710119403 CN101094406B (en) 2007-07-23 2007-07-23 Method and device for transferring video data stream

Country Status (1)

Country Link
CN (1) CN101094406B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2804375A1 (en) * 2013-02-22 2014-11-19 Thomson Licensing Coding and decoding methods of a picture block, corresponding devices and data stream
CN104363454B (en) * 2014-09-01 2017-10-27 北京大学 A high-rate video encoding and image decoding method and system
CN104639909A (en) * 2015-02-06 2015-05-20 达声蔚 Method and device for transmitting video

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1404302A (en) 2001-09-05 2003-03-19 北京中科大洋科技发展股份有限公司 TV program making system and method with double-bit rate video stream
US7082221B1 (en) 2000-09-29 2006-07-25 Intel Corporation Bandwidth determination for multiple layer digital video
CN1969562A (en) 2004-05-13 2007-05-23 高通股份有限公司 Synchronization of audio and video data in a wireless communication system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7082221B1 (en) 2000-09-29 2006-07-25 Intel Corporation Bandwidth determination for multiple layer digital video
CN1404302A (en) 2001-09-05 2003-03-19 北京中科大洋科技发展股份有限公司 TV program making system and method with double-bit rate video stream
CN1969562A (en) 2004-05-13 2007-05-23 高通股份有限公司 Synchronization of audio and video data in a wireless communication system

Also Published As

Publication number Publication date
CN101094406A (en) 2007-12-26

Similar Documents

Publication Publication Date Title
Keesman et al. Transcoding of MPEG bitstreams
KR101157058B1 (en) Encoding apparatus and encoding method
US8270475B2 (en) Method and apparatus for coding image information, method and apparatus for decoding image information, method and apparatus for coding and decoding image information, and system of coding and transmitting image information
US6879268B2 (en) Adaptive variable length coding of digital video
CN100380980C (en) Method and device for indicating quantizer parameters in a video coding system
CN1140130C (en) Quantization matrix for still and moving picture coding
US6510177B1 (en) System and method for layered video coding enhancement
CA2644605C (en) Video processing with scalability
CN1035977C (en) Apparatus for video data quantization control
CN1327395C (en) Method and system for context-based adaptive binary arithmetic coding
US20040136457A1 (en) Method and system for supercompression of compressed digital video
JP4138056B2 (en) Multi-standard decompression and / or compression device
US7170938B1 (en) Rate control method for video transcoding
US6031937A (en) Method and apparatus for video compression using block and wavelet techniques
KR101010722B1 (en) Coding transform coefficients in image/video encoder and/or decoders
RU2354073C2 (en) Conversion factor coding in image and videodata coders and/or decoders
US9762932B2 (en) System and method for transcoding entropy-coded bitstreams
US6677868B2 (en) Entropy coding with adaptive syntax to replace high probability symbols with lower probabilities symbols
JP3529624B2 (en) Encoding method and encoding apparatus of the picture
CA2160300C (en) Method and apparatus for partially recompressing digital signals
KR100880039B1 (en) Method and system for achieving coding gains in wavelet-based image codecs
CN1984329B (en) Device for coding, method for coding, system for decoding, method for decoding video data
US20060088098A1 (en) Method and arrangement for reducing the volume or rate of an encoded digital video bitstream
CN100592784C (en) Digital video signal processing apparatus of mobile communication system and method thereof
US8396307B2 (en) Image processing system, image processing method and program

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted
C17 Cessation of patent right