CN102065293B - Image compression method based on space domain predictive coding - Google Patents
Image compression method based on space domain predictive coding Download PDFInfo
- Publication number
- CN102065293B CN102065293B CN201010555095.5A CN201010555095A CN102065293B CN 102065293 B CN102065293 B CN 102065293B CN 201010555095 A CN201010555095 A CN 201010555095A CN 102065293 B CN102065293 B CN 102065293B
- Authority
- CN
- China
- Prior art keywords
- image
- mode
- prediction
- residual image
- coded macroblocks
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 230000006835 compression Effects 0.000 title abstract description 22
- 238000007906 compression Methods 0.000 title abstract description 22
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 5
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 5
- 230000003068 static effect Effects 0.000 abstract description 5
- 238000000638 solvent extraction Methods 0.000 abstract 5
- 230000000694 effects Effects 0.000 description 6
- 230000009466 transformation Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention provides an image compression method based on space domain predictive coding, comprising the following steps of: inputting a static image, and divide the image into a 16*16 coding macro block; partitioning according to multiple partitioning modes, and predicting a spatial domain; eliminating the redundant information of the image in the spatial domain with a prediction method comprising texture synthesis; predicting the coding macro block by multiple prediction modes comprising the texture synthesis mode so as to obtain a predicted image and a residual image; comparing to obtain a prediction mode with the minimum residual image in the partitioning mode; determining the optimal partitioning mode by a rate distortion function; coding by the prediction mode with the minimum residual image and the residual image in the optimal partitioning mode to serve as the code stream of the coding micro block; and obtaining the code stream of the left coding micro blocks. The image compression method can furthest reduce the redundant information of the image in the spatial domain and improve the compression ratio for compressing images.
Description
Technical field
The present invention relates to a kind of method for compressing image based on space domain predictive coding.
Background technology
It is a difficult point in image compression that thereby existing information how to utilize image reduces the amount of information raising compression ratio of image, because image is static, the redundancy of removal of images can only be from of the image itself, video sequence can be eliminated redundancy from time domain, classical method for compressing image is all to improve smoothed image in the rate of decay of frequency domain by the method for transition coding at present, as the dct transform in JPEG compression, the wavelet transformation of JPEG2000 etc., these methods all do not utilize the characteristic in image space territory to reduce the comentropy of image, the character of wavelet transformation local space-time two weight analysis of JEPG2000 wherein, compared with JPEG, be significantly improved.
The image of continuously smooth has good correlation in spatial domain, utilizes this character to predict spatial data, to the predicated error compressed image preferably of encoding.In the image of structural strong, good stability, the prediction of spatial domain has desirable compression effectiveness (for example compressing a smoothing prediction piece only needs the information of 4bit left and right to represent this piece, with block size 16 * 16, calculate, compression ratio can reach 512), and it is stronger for randomness, the region that texture is many, it is helpless that the coding of spatial domain seems.The compression ratio that how to utilize the characteristic of texture image further to improve image is the road of the certainty that improves of image compression performance.
Classical spatial prediction mode has obtained desirable effect in the compression of video compression standard I frame H.264, it is divided into image 16 * 16 macro block, for each, can also be subdivided into 8 * 8 or 4 * 4 size, this point-score can improve the granularity of piecemeal greatly, can make full use of the relevant information of various piecemeals.For each piecemeal, in spatial domain, can predict by DC, from top to bottom, from left to right, do simple linear prediction in 9 directions such as diagonally opposing corner.This algorithm can be obtained more satisfactory effect for changing more continuous and smooth region.But bad for sudden change and periodic effect data, Fig. 1, Fig. 2 are that they are for the prediction effect figure of these two kinds of signals.
From experimental result above, undesirable for the prediction effect of texture image, need to there is the predictive mode of good texture image.H.264 standard is used for compression of video sequence, is not used for specially processing static image.
Summary of the invention
The object of this invention is to provide a kind of method for compressing image based on space domain predictive coding, can solve the problem of the compression method that is not used for specially processing static image.
The object of the invention is to be achieved through the following technical solutions:
Propose a kind of method for compressing image based on space domain predictive coding, comprise the steps: the coded macroblocks of inputting still image and dividing 16 * 16 sizes; According to the go forward side by side row space territory prediction of multiple macroblock mode piecemeal, adopt and comprise that Forecasting Methodology removal of images that texture synthesizes is in the redundant information of spatial domain; Employing comprises that the multiple predictive mode of texture synthesis model predicts described coded macroblocks, to obtain predicted picture, residual image; Relatively obtain the predictive mode of residual image minimum under described macroblock mode; Percent of pass distortion function is determined best macroblock mode, adopts predictive mode and the residual image of residual image minimum under this best macroblock mode to encode, as the code stream of this coded macroblocks; Obtain the code stream of remaining coded macroblocks.
According to the go forward side by side row space territory prediction of multiple macroblock mode piecemeal, employing comprises that Forecasting Methodology removal of images that texture synthesizes, in the redundant information of spatial domain, also comprises: the described encoding block of encoding is to obtain spatial prediction pattern and the partitioned mode of all described encoding blocks.
Beneficial effect: utilize multiple prediction group syntype to process static image, improve the prediction effect of spatial domain, improve compression efficiency.By select best partitioned mode and best predictive mode for still image, thus reduce to greatest extent image in the redundant information of spatial domain to improve the compression ratio of image compression.
Accompanying drawing explanation
According to drawings and embodiments the present invention is described in further detail below.
Fig. 1 is one of design sketch carrying out in prior art image compression;
Fig. 2 be in prior art, carry out image compression design sketch two;
Fig. 3 is the design sketch that the present invention uses the synthetic predictive mode of texture to predict;
Fig. 4 is the flow chart of a kind of method for compressing image based on space domain predictive coding of the present invention;
Fig. 5 is the flow chart of a kind of method for compressing image based on space domain predictive coding of another scene of the present invention.
Embodiment
The present invention proposes a kind of method for compressing image based on space domain predictive coding, basic thought is: by select best partitioned mode and best predictive mode for still image, thereby reduce to greatest extent image in the redundant information of spatial domain to improve the compression ratio of image compression.
Embodiment 1
A method for compressing image based on space domain predictive coding, comprises the steps:
S11, input still image are also divided the coded macroblocks of 16 * 16 sizes;
S12, according to the go forward side by side row space territory prediction of multiple macroblock mode piecemeal, adopt and comprise that Forecasting Methodology removal of images that texture synthesizes is in the redundant information of spatial domain;
S13, adopt and to comprise that the multiple predictive mode of texture synthesis model predicts coded macroblocks, to obtain predicted picture, residual image;
S14, relatively obtain the predictive mode of residual image minimum under macroblock mode;
S15, percent of pass distortion function are determined best macroblock mode, adopt predictive mode and the residual image of residual image minimum under this best macroblock mode to encode, as the code stream of this coded macroblocks;
S15, obtain the code stream of remaining coded macroblocks.
Embodiment 2
A method for compressing image based on space domain predictive coding, comprising:
S21, input still image, and this still image is divided into the coded macroblocks of 16 * 16 sizes.
S22, select one of them coded macroblocks, adopt different partitioned modes to carry out piecemeal to this coded macroblocks.
S221, coded macroblocks is divided into the encoding block of 48 * 8, adopts ten kinds of predictive modes to obtain one by one original image, predicted picture and the residual image under this predictive mode.
Wherein, ten kinds of predictive modes comprise: nine linear prediction patterns and a synthetic predictive mode of texture, nine linear prediction patterns comprise: horizontal forecast, 8 directional prediction modes such as vertical prediction and lower-left prediction and mean prediction (DC prediction), wherein, the synthetic predictive mode of texture is specially, encoding block to be predicted is divided into several fritters of 2 * 2, getting upper left 60 pixels of each 2 * 2 fritter is template, a given hunting zone (being conventionally taken as between 8 to 12), adopt the method for piece coupling in decoded image, finds with work as front template apart from the image block of 3 or 58 * 8 sizes of minimum as candidate blocks.After being weighted on average to the pixel value in the lower right corner in candidate blocks 2 * 2, think the predicted value to pixel value in current 2 * 2 fritters to be predicted.
S2211, each encoding block is encoded, obtain the pixel value of each encoding block, and by the pixel value of encoding block, obtain the original image of encoding block, the original image of calculation code macro block.
S2212, gather the pixel value of first encoding block, adopt the pixel value of second encoding block of a kind of predictive mode prediction in ten kinds of predictive modes, and the pixel value of second encoding block obtaining according to prediction obtains the predicted picture of second encoding block.
S2213, obtain the predicted picture of this coded macroblocks integral body, according to the original image of the predicted picture of this coded macroblocks integral body and this coded macroblocks integral body, calculate the residual image of this coded macroblocks integral body.
S2214, adopt ten kinds of predictive modes to obtain ten groups of predicted pictures and residual image and the predictive mode of this coded macroblocks respectively.
S2215, one group of predicted picture preserving residual image minimum wherein and residual image and predictive mode.
S222, coded macroblocks is divided into 84 * 4 encoding blocks, adopts ten kinds of predictive modes to obtain one by one original image, predicted picture and the residual image under this predictive mode.
Wherein, ten kinds of predictive modes comprise: nine linear prediction patterns and a synthetic predictive mode of texture, nine linear prediction patterns comprise: horizontal forecast, 8 directional prediction modes such as vertical prediction and lower-left prediction and mean prediction (DC prediction), wherein, the synthetic predictive mode of texture is specially, encoding block to be predicted is divided into several fritters of 2 * 2, getting upper left 60 pixels of each 2 * 2 fritter is template, a given hunting zone (being conventionally taken as between 8 to 12), adopt the method for piece coupling in decoded image, finds with work as front template apart from the image block of 3 or 58 * 8 sizes of minimum as candidate blocks.After being weighted on average to the pixel value in the lower right corner in candidate blocks 2 * 2, think the predicted value to pixel value in current 2 * 2 fritters to be predicted.
S2221, each encoding block is encoded, obtain the pixel value of each encoding block, and by the pixel value of encoding block, obtain the original image of encoding block, the original image of calculation code macro block.
S2222, gather the pixel value of first encoding block, adopt the pixel value of second encoding block of a kind of predictive mode prediction in ten kinds of predictive modes, and the pixel value of second encoding block obtaining according to prediction obtains the predicted picture of second encoding block.
S2223, obtain the predicted picture of this coded macroblocks integral body, according to the original image of the predicted picture of this coded macroblocks integral body and this coded macroblocks integral body, calculate the residual image of this coded macroblocks integral body.
S2224, adopt ten kinds of predictive modes to obtain ten groups of predicted pictures and residual image and the predictive mode of this coded macroblocks respectively.
S2225, one group of predicted picture preserving residual image minimum wherein and residual image and predictive mode.
S23, percent of pass distortion function obtain best partitioned mode, and preserve one group of predicted picture, residual image and predictive mode of residual image minimum under this partitioned mode.
The residual image that residual image deducts after quantification treatment obtains distortion, residual image after coded quantization obtains code check, code check and distortion are inversely proportional to, the larger distortion of code check is less, the less distortion of code check is larger, and percent of pass distortion function is determined much distortions under how many code checks, and effect is better, thereby select best partitioned mode, and preserve one group of predicted picture, residual image and predictive mode under best partitioned mode.
S24, above-mentioned residual image and predictive mode are encoded, obtain the final code stream of this coded macroblocks.
Be specially: by dct transform, obtain dct transform coefficient, dct transform coefficient is carried out to the dct transform coefficient after adaptive quantizing is quantized, adopt entropy coder to carry out arithmetic coding to the dct transform coefficient after quantizing and obtain code stream, and by this code stream and the code stream merging that adopts entropy coder that predictive mode is encoded and obtained, generate final binary code stream output as the final code stream of this coded macroblocks.
On the other hand, carry out inverse dct transform, the residual image after inverse transformation and original predicted picture are added as the image of having encoded, for the prediction to next coded macroblocks pixel value.
S25, select other coded macroblocks to encode, until all coded macroblockss of encoding obtain the final code stream of whole image.
The present invention is not limited thereto execution mode, take inventive concept as basic related realization assembly is all in protection scope of the present invention.
Claims (2)
1. the method for compressing image based on space domain predictive coding, is characterized in that, comprises the steps:
Input still image is also divided the coded macroblocks of 16 * 16 sizes;
According to the go forward side by side row space territory prediction of multiple macroblock mode piecemeal, adopt and comprise that Forecasting Methodology removal of images that texture synthesizes is in the redundant information of spatial domain;
Employing comprises that the multiple predictive mode of texture synthesis model predicts described coded macroblocks, to obtain predicted picture, residual image;
Relatively obtain the predictive mode of residual image minimum under described macroblock mode;
Percent of pass distortion function is determined best macroblock mode, adopts predictive mode and the residual image of residual image minimum under this best macroblock mode to encode, as the code stream of this coded macroblocks;
Obtain the code stream of remaining coded macroblocks.
2. the method for claim 1, is characterized in that, according to the go forward side by side row space territory prediction of multiple macroblock mode piecemeal, adopts and comprises that Forecasting Methodology removal of images that texture synthesizes, in the redundant information of spatial domain, also comprises:
Encode described encoding block to obtain spatial prediction pattern and the partitioned mode of all described encoding blocks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010555095.5A CN102065293B (en) | 2010-11-23 | 2010-11-23 | Image compression method based on space domain predictive coding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010555095.5A CN102065293B (en) | 2010-11-23 | 2010-11-23 | Image compression method based on space domain predictive coding |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102065293A CN102065293A (en) | 2011-05-18 |
| CN102065293B true CN102065293B (en) | 2014-07-16 |
Family
ID=44000362
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010555095.5A Active CN102065293B (en) | 2010-11-23 | 2010-11-23 | Image compression method based on space domain predictive coding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102065293B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102333220B (en) * | 2011-10-21 | 2013-11-06 | 北京大学 | Video coding and decoding method capable of selectively finishing predictive coding in transform domain |
| CN102622198A (en) * | 2012-02-29 | 2012-08-01 | 中山大学 | Method and system for visualizing data |
| CN113570677A (en) * | 2017-05-03 | 2021-10-29 | 西安万像电子科技有限公司 | Image compression method and device |
| CN112055223A (en) * | 2020-08-21 | 2020-12-08 | 浙江大华技术股份有限公司 | Image coding and decoding method and coder-decoder |
| CN112702598B (en) * | 2020-12-03 | 2024-06-04 | 浙江智慧视频安防创新中心有限公司 | Method, device, electronic equipment and medium for encoding and decoding based on displacement operation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1615020A (en) * | 2004-11-10 | 2005-05-11 | 华中科技大学 | Method for pridicting sortable complex in frame |
| CN101141647A (en) * | 2007-08-24 | 2008-03-12 | 上海广电（集团）有限公司中央研究院 | AVS video coding based fast intraframe predicting mode selecting method |
| CN101433094A (en) * | 2006-04-28 | 2009-05-13 | 株式会社Ntt都科摩 | Image predictive coding device, image predictive coding method, image predictive coding program, image predictive decoding device, image predictive decoding method and image predictive decoding progra |
-
2010
- 2010-11-23 CN CN201010555095.5A patent/CN102065293B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1615020A (en) * | 2004-11-10 | 2005-05-11 | 华中科技大学 | Method for pridicting sortable complex in frame |
| CN101433094A (en) * | 2006-04-28 | 2009-05-13 | 株式会社Ntt都科摩 | Image predictive coding device, image predictive coding method, image predictive coding program, image predictive decoding device, image predictive decoding method and image predictive decoding progra |
| CN101141647A (en) * | 2007-08-24 | 2008-03-12 | 上海广电（集团）有限公司中央研究院 | AVS video coding based fast intraframe predicting mode selecting method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102065293A (en) | 2011-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104967850B (en) | The method and apparatus that image is coded and decoded by using big converter unit | |
| CN100579234C (en) | Image encoding/decoding method and apparatus therefor | |
| CN102763411B (en) | Method and apparatus to Video coding and the method and apparatus to video decode | |
| CN100361536C (en) | Device and method for predicting and encoding image and recording medium | |
| EP3107292B1 (en) | Video encoding method and apparatus, and video decoding apparatus | |
| CN104796710B (en) | The method decoded to image | |
| CN101854548B (en) | Wireless multimedia sensor network-oriented video compression method | |
| CN104094600B (en) | For the Video coding and the method and apparatus of decoding based on hierarchical data unit predicted including quantization parameter | |
| CN103248895B (en) | A kind of quick mode method of estimation for HEVC intraframe coding | |
| CN101779463B (en) | Method for processing images and the corresponding electronic device | |
| KR20100004037A (en) | Method and apparatus for encoding video, and method and apparatus for decoding video | |
| CN103024383A (en) | Intra-frame lossless compression coding method based on HEVC (high efficiency video coding) frame | |
| KR20110112224A (en) | Method and apparatus for encdoing/decoding information regarding encoding mode | |
| MX2012000614A (en) | Method and apparatus for encoding and decoding image by using large transformation unit. | |
| CN103634606A (en) | Video encoding method and apparatus | |
| CN103596004A (en) | Intra-frame prediction method and device based on mathematical statistics and classification training in HEVC | |
| CN102065293B (en) | Image compression method based on space domain predictive coding | |
| KR20090087471A (en) | Method of encoding an image and device implementing said method | |
| CN104702959B (en) | A kind of intra-frame prediction method and system of Video coding | |
| KR101512643B1 (en) | 2 Video encoding apparatus and Apparatus and Method of 2 dimensional ordering transform for image signal and Recording Medium therefor | |
| CN100539695C (en) | Image forecasting encoding/decoding apparatus and method and recording medium | |
| CN102045554A (en) | Airspace prediction method and device of video and image | |
| CN105025300B (en) | The method and apparatus encoded/decoded to image | |
| CN100477793C (en) | Method and device for predicting discrete cosine coefficient | |
| CN100512445C (en) | Predictive coding method among interlacing video sequence frame internal fields |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: An image compression method based on spatial domain predictive coding Granted publication date: 20140716 Pledgee: Wuxi Xishan sub branch of Bank of China Ltd. Pledgor: WUXI GANGWAN NETWORK TECHNOLOGY Co.,Ltd. Registration number: Y2024980036088 |