专利摘要:
The present invention provides a difference image for determining an optimal block by using a block boundary value in a spatial domain when restoring a bit error of difference image (MCPE) information transmitted in a predetermined block unit in an image decoding system. A bit error restoring apparatus, comprising: a video decoding system for restoring bitstream data of a difference image (MCPE) signal transmitted in a structure including parity bits in a transport block unit, A bit stream generation unit configured to generate bit streams having a hamming distance of 1 as candidate bit streams for the bit stream data; varying each candidate bit stream transmitted from the candidate bit stream generation unit; Candidate block generation for generating candidate blocks corresponding to candidate bit strings by length decoding, inverse quantization, and inverse discrete cosine transform processing An adder configured to add motion vector information corresponding to the difference image signal currently transmitted to respective candidate blocks output from the candidate block generator; each candidate block to which motion vector information transmitted from the adder is added and present is added; Bit error for detecting the stiffness of the boundary region between blocks adjacent to the block position of the transmitted difference image signal, and determining and outputting the candidate blocks having the most gentle neighboring blocks and the boundary region as the difference image block in which the bit error is restored. It is configured to include a recovery unit.
公开号:KR19980053573A
申请号:KR1019960072679
申请日:1996-12-27
公开日:1998-09-25
发明作者:이상훈
申请人:배순훈;대우전자 주식회사;
IPC主号:
专利说明:

Bit Error Restoration Device for Differential Image in Image Decoding System
The present invention relates to a bit error restoring apparatus for motion image prediction (Motion Compensation Predict Error, MCPE) in an image decoding system. In particular, one parity bit per bitstream of an MCPE transmitted in a predetermined block unit The present invention relates to a bit error restoring apparatus of an MCPE for restoring bit errors of an MCPE bitstream in an image decoding system for restoring additionally transmitted bitstream data.
In the video signal processing system that transmits data in units of blocks, the MCPE information includes a difference value between pixel data in a currently applied block and pixel data in a block at the same position in a previous frame, as shown in FIG. 1. That is, when the pixel data in the (i, j) -th block of the current frame is configured as A, and the pixel data in the (i, j) -th block in the previous frame is configured as A ', as shown in FIG. Likewise, MCPE information is organized. Therefore, the blocks of the MCPE information transmitted have the same size as the transmission block unit of the general video signal.
In order to detect the bit error of the MCPE information configured as described above, the bitstream data of the MCPE transmission block unit including the video encoding system and the parity bit transmitted by adding a parity bit of 1 bit in MCPE transmission block unit is transmitted. An image decoding system has been proposed that checks whether a bit error occurs depending on whether the added value is even or odd.
In particular, if the proposed image decoding system determines that a bit error is generated by the above-described method, it is possible to recover the bit error to the bit strings having a hamming distance of 1 for MCPE bitstream data in units of blocks except parity bits. Candidate blocks, motion vector information is added to the generated candidate blocks, and candidate blocks to which motion vectors are added and blocks adjacent to MCPE information blocks in which bit errors are generated (top block and left block). A bit error recovery apparatus is implemented in such a manner that a degree of correlation between adjacent blocks is detected and a block having the highest correlation with an adjacent block is detected as a block in which a bit error is restored.
However, the above-described bit error recovery apparatus has a problem in that it is difficult to perform real-time processing since the correlation between the generated candidate blocks and adjacent blocks is detected in the frequency domain when detecting the block in which the bit error is restored.
Accordingly, the present invention has been made to solve the above-described drawbacks. In the image decoding system, when a bit error is restored for differential image (MCPE) information transmitted in a predetermined block unit, a block boundary value in a spatial domain is provided. It is an object of the present invention to provide a differential image bit error restoration apparatus for determining an optimal block by using.
In order to achieve the above object, the differential image bit error restoration apparatus according to the present invention is an image decoding system for restoring bitstream data of a differential image (MCPE) signal transmitted in a structure including a parity bit in a transport block unit, If a bit error is detected in the bitstream data of the currently transmitted difference image signal, a candidate bit string generator for generating a bit stream having a hamming distance of 1 as a candidate bit string with respect to the bit stream data; A candidate block generation unit configured to perform variable length decoding, inverse quantization, and inverse discrete cosine transformation on each candidate bit stream to generate candidate blocks corresponding to the candidate bit stream; currently transmitted to respective candidate blocks output from the candidate block generator An adder configured to add motion vector information corresponding to the difference image signal; a motion vector transmitted from the adder The bit error is reconstructed by detecting the gentleness of the boundary region between each candidate block to which information is added and blocks adjacent to the block position of the currently transmitted difference image signal. And a bit error reconstruction unit for determining and outputting the difference image block.
1 is a diagram illustrating a general difference image detection
2 is a functional block diagram of an image decoding system having a bit error restoration apparatus for a difference image according to the present invention.
3 is a detailed functional block diagram of a bit error recovery unit shown in FIG.
4A and 4B are exemplary diagrams illustrating the relationship between blocks for explaining the operation of the sidematch tester shown in FIG.
Explanation of symbols for main parts of the drawings
200: bit error detection unit 210: switch
220; Variable Length Decoder
230: bit error recovery device of the difference image
231: Candidate bit string generation unit 232: Candidate block generation means
233: bit error recovery unit
The above and other objects and various advantages of the present invention will become more apparent from the preferred embodiments of the invention described below with reference to the accompanying drawings by those skilled in the art.
Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.
FIG. 2 is a block diagram of an image decoding system including a bit error restoration apparatus of a difference image (hereinafter, referred to as MCPE) according to the present invention. When bit stream data of MCPE information in a predetermined block unit including parity bits is applied, bit error occurs. Bit error detection unit 200 for detecting the occurrence of occurrence, variable length decoding unit 220 for variable length decoding the bitstream data of the MCPE information applied when the bits of the transmitted MCPE information is normal (hereinafter VLD (Variable Length Decoder) )), The switching state is controlled by the differential image bit error recovery apparatus 230 and the bit error detection unit 200 for restoring the bit error of the MCPE information applied when the transmitted MCPE information bit is abnormal The switch 210 performs transmission processing to the above-described VLD 220 or the differential image bit error recovery apparatus 230 for the applied MCPE information.
In particular, the differential image bit error recovery apparatus 230 may include a candidate bit string generator 231 and a candidate bit string generator 231 for generating candidate bit strings having a Hamming distance of 1 for the applied bit string. The candidate block generation means 232 for generating candidate blocks corresponding to the column, and the optimum candidate block is selected according to the degree of boundary region matching between the candidate blocks provided from the candidate block generation means 232 and the adjacent block in the spatial region. And a bit error recovery unit 235 for restoring the bit error.
The video decoding system configured as described above is operated as follows.
When the MCPE signal of a predetermined block unit including the parity bit is applied, the bit error detection unit 200 and the switch 210 are applied.
The bit error detection unit 200 adds a bitstream of the MCPE information including the parity bit, and determines whether a bit error occurs according to whether the added result value is even or odd. As a result of the determination, when no bit error occurs, the bit error detection unit 200 outputs a switching control signal so that the switch 210 is switched to the a1 contact.
Accordingly, the switch 210 transmits the MCPE information applied through the reference contact point to the VLD 220.
When the MCPE information including the parity bit is applied from the switch 220, the VLD 220 decodes and outputs the variable length, as is well known, for the MCPE information excluding the parity bit.
On the other hand, if a bit error is detected in the transmitted MCPE information as a result of the bit error detection unit 200, the switch 210 is switched to the a2 contact to restore the MCPE information including the parity bit applied to the difference image bit error Transmit to device 230.
The differential image bit error recovery apparatus 230 transmits the MCPE information including the parity bit to the candidate bit string generator 231 when the MCPE information including the parity bit is applied. The candidate bit string generator 231 generates and outputs bit strings having a hamming distance of 1 as candidate bit strings. The output candidate bit strings are transmitted to the candidate block generating means 232.
The candidate block generation unit 232 includes a candidate block generation unit 233 and an adder 234, and when each candidate bit string is applied, the candidate block generation unit 232 transmits the candidate block generation unit 233 to the candidate block generation unit 233.
The candidate block generation unit 233 performs variable length decoding, inverse quantization (IQ), and inverse discrete cosine transform (IDCT) processing on each candidate bit string applied to form and output a candidate block corresponding to the candidate bit string. Each candidate block output is transmitted to the adder 234.
The adder 234 adds the motion vector information corresponding to the currently applied MCPE information to each candidate block subjected to inverse discrete cosine transform, and outputs the added motion vector information. Each candidate block to which the output motion vector information is added is transmitted to the bit error recovery unit 235.
As shown in FIG. 3, the bit error recovery unit 235 includes a side match tester 300 and a memory 310, and intersects blocks adjacent to the MCPE image block where the current bit error has occurred. The degree of gentleness of the boundary region is detected, and candidate blocks with the most gentle boundary region are output as reconstructed MCPE information.
That is, when candidate blocks (IDCT candidate difference image + motion vector information) generated from the candidate block generation unit 232 are transmitted, the candidate blocks are applied to the side match tester 300. The sidematch tester 300 detects the difference between the candidate block and the adjacent block by using pixel data located in the boundary region of each block adjacent to the MCPE block position where the current error occurs, and the candidate having the smallest difference. Determine the block as the best block and output it.
In more detail, as shown in FIG. 4A, when candidate blocks of the MCPE block in which the current error has occurred are transmitted from the candidate block generating means 232 described above, the side match test unit 300 may be configured as illustrated in FIG. 4B. For each of the applied candidate blocks, X1 to X8 pixel data of the highest horizontal line in contact with the lower boundary plane of the top block adjacent to the MCPE block in which the error occurred are extracted, and the adjacent left block is extracted. Extract Y1 ~ Y8 pixel data of the leftmost vertical line on the right boundary. For the top block provided from the memory 310, the X1 'to X8 pixel data of the lowest horizontal line in contact with the upper boundary plane of the MCPE block in which the error is generated are extracted, and the adjacent left block of the MCPE block in which the error occurs is extracted. Pixel data of Y1 'to Y8' of the rightmost vertical line in contact with the left boundary is extracted, and the boundary difference value Dsm between adjacent blocks for candidate blocks of each MCPE is obtained as shown in Equation 1 below.
[Equation 1]
here Is the pixel data located on the top horizontal line adjacent to the lower boundary of the top block in the candidate block. Is the pixel data located on the lowest horizontal line in the top block that is adjacent to the upper boundary of the current block. Is the pixel data located on the leftmost vertical line adjacent to the right boundary of the left block in the candidate block. Is the pixel data located on the rightmost vertical line adjacent to the left boundary of the current block in the adjacent left block.
As described above, after all boundary difference values (Dsm) with neighboring blocks for each MCPE candidate block are obtained, the candidate block having the smallest difference value is detected and output as an optimal block that has recovered the bit error. The data output at this time becomes pixel data expressed in the spatial region as described above.
The memory 310 stores a predetermined number of block data so as to provide the side match tester 300 with pixel data of the top block adjacent to the upper side and the block adjacent to the left side of the MCPE information block in which an error occurs, and in one block unit. The stored data is updated.
As described above, according to the present invention, a space between a candidate block and an adjacent block generated to recover a bit error of MCPE information transmitted in a predetermined block unit including a parity bit in an image decoding system decoding in a predetermined block unit By providing a bit error restoring device for the difference image that determines the optimal candidate block in which the bit error is restored by using the boundary value in the area, the correlation detection method in the conventional frequency domain is used in determining the optimal block. There is an advantage to real time processing.
It will be appreciated by those skilled in the art that the present invention is not limited to the above-described embodiments and that various changes and modifications can be made without departing from the spirit of the present invention. Therefore, the technical scope of the present invention should not be limited to the contents described in the detailed description of the specification but should be defined by the claims.
权利要求:
Claims (2)
[1" claim-type="Currently amended] An image decoding system for restoring bitstream data of a difference image (MCPE) signal transmitted in a structure including a parity bit in a transport block unit,
A candidate bit string generator 231 for generating bit strings having a hamming distance of 1 as candidate bit strings when bit errors are detected in bitstream data of the differential video signal currently transmitted;
A candidate block generator 233 for variable length decoding, inverse quantization, and inverse discrete cosine transform processing of the candidate bit strings transmitted from the candidate bit string generator 231 to generate candidate blocks corresponding to the candidate bit strings;
An adder 234 for adding motion vector information corresponding to the currently transmitted difference image signal to respective candidate blocks output from the candidate block generator 233;
Degradation of the boundary area between each candidate block to which motion vector information transmitted from the adder 234 is added and blocks adjacent to the block position of the currently transmitted difference image signal is detected to detect the adjacent blocks and the boundary area. And a bit error reconstruction unit (235) for determining and outputting the most gentle candidate block as a differential image block in which the bit error is reconstructed.
[2" claim-type="Currently amended] The block error recovery unit 235 of claim 1, wherein the bit error recovery unit 235 blocks the adjacent blocks to a block position (TOP block) and a block adjacent to the left (Left block) to the upper side of the block position of the currently transmitted difference image signal. And a severity of a boundary area between the respective candidate blocks and adjacent blocks.
类似技术:
公开号 | 公开日 | 专利标题
RU2719433C1|2020-04-17|Device for encoding images with prediction, device for decoding images with prediction, method of encoding images with prediction, method of decoding images with prediction, program for encoding images with prediction and program for decoding images with prediction
US9800889B2|2017-10-24|Video decoding device, video decoding method, video decoding program, and video decoding integrated circuit
US8599923B2|2013-12-03|Motion vector encoding device and decoding device
JP5642859B2|2014-12-17|Method and apparatus for encoding motion information
US6128342A|2000-10-03|Video coding apparatus which outputs a code string having a plurality of components which are arranged in a descending order of importance
US8098940B2|2012-01-17|Method of and apparatus for encoding signals, and method of and apparatus for decoding the encoded signals
DE69724672T2|2004-07-15|Video coding method for coding shape and texture signals using different modes
ES2295895T3|2008-04-16|Procedure, device and system for coding and decoding side information for multimedia transmission.
KR100328808B1|2002-03-14|A method for inserting and detecting electronic watermark data into a digital image and a device for the same
US5086439A|1992-02-04|Encoding/decoding system utilizing local properties
CN101268701B|2012-06-20|Method for execution by electronic processor and device therefor
US8009923B2|2011-08-30|Method and system for motion estimation with multiple vector candidates
KR100382440B1|2003-05-09|Digital image encoding and decoding method and digital image encoding and decoding device using the same
EP0719049B1|2003-05-28|Method and apparatus for image coding
JP2013070399A|2013-04-18|Scalable video coding with filtering of lower layers
KR0148154B1|1998-09-15|Coding method and apparatus with motion dimensions
JP4034380B2|2008-01-16|Image encoding / decoding method and apparatus
JP3847827B2|2006-11-22|Motion vector detection method
TWI301953B|2008-10-11|Methods and apparatuses for video encoding
KR100603175B1|2006-07-24|Video encoding method, video decoding method, video encoding apparatus, video decoding apparatus, a computer readable recording medium having a video encoding program recorded thereon, and a computer readable recording medium having a video decoding program recorded thereon
KR101366242B1|2014-02-20|Method for encoding and decoding motion model parameter, and method and apparatus for video encoding and decoding using motion model parameter
US8254456B2|2012-08-28|Method and apparatus for encoding video and method and apparatus for decoding video
KR100413153B1|2003-12-31|Picture encoder and picture decoder
CN101193302B|2010-09-29|Illumination compensation method and apparatus and video encoding and decoding method and apparatus
JP3169783B2|2001-05-28|Video encoding / decoding system
同族专利:
公开号 | 公开日
KR100212556B1|1999-08-02|
引用文献:
公开号 | 申请日 | 公开日 | 申请人 | 专利标题
法律状态:
1996-12-27|Application filed by 배순훈, 대우전자 주식회사
1996-12-27|Priority to KR1019960072679A
1998-09-25|Publication of KR19980053573A
1999-08-02|Application granted
1999-08-02|Publication of KR100212556B1
优先权:
申请号 | 申请日 | 专利标题
KR1019960072679A|KR100212556B1|1996-12-27|1996-12-27|Apparatus for reconstructing bits error about mcpe in the image decoder|
[返回顶部]