CN103338363A - System and method for video compression sensing and encoding based on measurement domain block classification - Google Patents
System and method for video compression sensing and encoding based on measurement domain block classification Download PDFInfo
- Publication number
- CN103338363A CN103338363A CN2013100892358A CN201310089235A CN103338363A CN 103338363 A CN103338363 A CN 103338363A CN 2013100892358 A CN2013100892358 A CN 2013100892358A CN 201310089235 A CN201310089235 A CN 201310089235A CN 103338363 A CN103338363 A CN 103338363A
- Authority
- CN
- China
- Prior art keywords
- territory
- video
- measuring
- measurement
- piece
- 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.)
- Granted
Links
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention relates to a system and a method for video compression sensing and encoding based on measurement of domain block classification. The system comprises a block compression sampling module, a block classification module of a measurement domain, a sampling rate allocation module, a sampling number allocation and secondary measurement module and a reconstruction module. An encoding end divides an input video block into sub-blocks which are equal in size and not mutually overlapped, and each of the sub-blocks is measured by using the same random measurement matrix to obtain measurement domain signals; the block classification module of the measurement domain uses the measurement domain signals as row vectors of the matrix to obtain a measurement domain characteristic image, a correlation relation model between the measurement domain signals and frequency domain signals is built, judgment of the block classification is carried out on the measurement domain characteristic image according to the relation model, sampling rates are allocated to different types of video blocks, the sampling number allocation and secondary measurement module determines a measurement number and a measurement matrix according to the sampling rates and the video blocks, video signals are measured again by using the measurement matrix to obtain measurement signals, and the measurement signals are transmitted to the reconstruction module to reconstruct.
Description
Technical field
The invention belongs to technical field of video coding, relate generally to video coding system and method based on compressed sensing, specifically is a kind of based on video compression perceptual coding system and the method for measuring the territory block sort, is used for video coding system.
Background technology
In normal video coded system H.264, adopt estimation to remove the temporal correlation of vision signal, adopt dct transform to remove spatial coherence, in fact further discover, no matter be still to have stronger correlation between the interior piece of frame or the DCT coefficient of interframe block, this correlation is strengthened along with the closing on of locus of piece, and it is relevant that we are called frequency domain with this correlation.
Utilize digital signal after the compressed sensing technology directly changes into the light signal of simulation compression based on the signal collecting device of compressed sensing, therefore in coding side original image prime field vision signal non-availability.Estimation in traditional video encoding standard, motion compensation, Predicting Technique etc. can't be applicable in the video coding system based on compressed sensing, therefore need study based on the signal correlation analytical method of measuring on the territory at coding side, to improve the compression efficiency of whole system.Suppose that perhaps coding side can access original pixel domain digital video signal, direct analytic signal correlation on pixel domain, and this has obviously run counter to the compressed sensing theory; Perhaps introduce restructing algorithm at coding side, analyze its correlation based on the pixel-domain video signal that reconstruct obtains, greatly increased the coding side complexity.Therefore, it is very difficult to utilize low complexity algorithm to obtain the pixel domain feature of picture signal at coding side.
Document " Zhirong Gao; Chengyi Xiong; Cheng Zhou and Hanxin Wang; Compressive Sampling with Coefficients Random Permutations for Image Compression; International Conference on Multimedia and Signal Processing (CMSP); pp.321-324,2011 " mention different image blocks in and have different degree of rarefications and complexity at transform domain; smooth block is with respect to texture block and edge block correspondence summation about non-zero DCT coefficients still less; the amplitude of texture block DCT coefficient changes milder, the amplitude of edge block DCT coefficient changes more violent.That is to say that the frequency-region signal feature can effectively reflect the pixel domain feature of image.
By top analysis as can be known, in the video coding system based on compressed sensing, by the relation between research measurement territory signal correlation and the corresponding frequency domain signal correlation, directly utilize and measure the correlation research that the territory signal is realized frequency-region signal, have very important researching value.
In the video coding system based on compressed sensing, usually frame of video being carried out piecemeal for the complexity that reduces memory space and reconstruct handles, and all video pieces or image block adopt identical processing method, and have ignored the characteristics of dissimilar video pieces itself.In the application of most of video communications, people's major concern be the subjective quality of image, rather than objective quality.Consider the characteristics of the vision system of human eye in addition, to the susceptibility of the zones of different human eye of image difference to some extent, utilize the block sort method that image is divided into different zones, then the Processing Algorithm different to different area applications.
Existing block sort method utilizes the characteristic distributions of mean variance on the pixel domain or frequency-region signal as classification foundation usually, and based on coding side pixel domain signal and the frequency-region signal non-availability of the video coding system of compressed sensing, existing block sort method is unavailable, therefore the block sort method on the research measurement territory in based on the video coding system of compressed sensing is carried out the distortion performance that different processing will help to improve system at dissimilar video pieces.
Summary of the invention
The objective of the invention is to overcome the shortcoming of above-mentioned prior art, propose a kind of based on video compression perceptual coding system and the method for measuring the territory block sort, at the characteristics of dissimilar video pieces dissimilar video pieces is carried out in various degree compression, to improve the distortion performance of whole system.
For achieving the above object, the invention provides a kind of video compression perceptual coding system based on measurement territory block sort, comprise that piecemeal compression sampling module, the block sort module of measuring the territory, sample rate distribution module, hits distribute and secondary measurement module and reconstructed module; Piecemeal compression sampling module is divided into the sub-piece of equal and opposite in direction non-overlapping copies with the input video piece, each sub-piece is measured with identical random measurement matrix measure the territory signal; Measure the block sort module in territory and measure the characteristic of field image module by extracting, measuring territory and frequency domain relational model module and judging module constitutes, the measurement territory signal of the sub-piece that the piecemeal compression sampling is obtained obtains measuring the characteristic of field image as the capable vector of matrix respectively, measure territory and frequency domain relational model module and set up the linear approximate relationship of measuring the correlation of territory signal and corresponding frequency-region signal, judging module is that theoretical foundation utilization measurement characteristic of field image carries out the judgement of piece classification with the correlative relationship model of measuring territory and frequency-region signal, be specially and calculate the cross covariance matrix of measuring the characteristic of field image, and drawn the variance of cross covariance entry of a matrix element by the correlative relationship model for the feature of weighing the video piece, realize the type judgement of video piece according to the size of variance yields, by judging module the video piece of input is judged to edge block, texture block or smooth block; System distributes different sample rates at the characteristics of court verdict and dissimilar video piece itself to dissimilar video pieces then; Hits distributes and the secondary measurement module is determined to measure number and measured matrix according to sample rate and video block sizes, utilizes to measure matrix and the video piece is measured again measure the territory signal, will measure the territory signal at last and be transferred to reconstructed module and be reconstructed.
Prior art is come the video piece is classified by obtain original image prime field signal in coding side reconstruct, increased the complexity of coding side greatly, suppose that perhaps coding side has original image prime field signal, run counter to the original intention of compressed sensing theory, and the present invention has set up measurement territory signal and discrete cosine transform (Discrete Cosine Transformation, DCT) the correlative relationship model of territory signal, and utilize this model to realize block sort in the measurement territory, carry out sample rate at the judgement type and distribute, improved the distortion performance of whole system.
Realization of the present invention is that also described foundation measures territory and the correlation of frequency domain relational model with two signals of cross covariance measurement,, set up and measured the correlation of territory signal and the linear approximate relationship model between the corresponding frequency domain signal correlation through the coefficient characteristic distributions after the discrete cosine transform according to vision signal.
The present invention utilizes and measures the correlation analysis that the territory signal has been realized corresponding DCT territory signal, for the video block sort of measuring on the territory provides theoretical foundation.
It is a kind of based on the video compression perceptual coding method of measuring the territory block sort that the present invention also provides, and the system that this method was suitable for is the video compression perceptual coding system based on measurement territory block sort that the present invention proposes, and this method comprises the steps:
Step 1. is with input picture piece or video piece x
bThe sub-piece that is divided into the equal and opposite in direction non-overlapping copies supposes to be divided into four sub-piece x
B1, x
B2, x
B3, x
B4, corresponding DCT territory signal is θ
1, θ
2, θ
3, θ
4, each sub-piece is measured matrix with the gaussian random of identical size measures measurement territory signal y respectively
1, y
2, y
3, y
4, measured value is the column vector of m * 1;
Step 2. is measured characteristic of field image y=[y with the measurement territory signal of sub-piece as the row vector structure of matrix
1, y
2, y
3, y
4]
T
Step 3. is set up the correlative relationship model of measuring territory signal and frequency-region signal:
Wherein, C
yFor measuring the cross covariance matrix on the territory, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory;
Step 4. pair is measured each element of territory covariance matrix and is asked variance, is got by the relational model of step 3:
Wherein, variance, C are asked in var () expression
yFor measuring the cross covariance matrix of characteristic of field image, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory.
Step 5. is adjudicated the type of video piece by following criterion, is judged to smooth block, edge block or texture block according to the big young pathbreaker's video of the variance piece of the cross covariance matrix element of measuring the characteristic of field image:
Wherein if be " if " the meaning, T
1, T
2Be empirical value.
Step 6. is carried out the sample rate distribution according to the video block type of judging module output, edge block＞texture block＞smooth block is satisfied in the distribution of sample rate, determine hits and measure matrix according to the size of sample rate and video piece, utilize this measurement matrix that current video block is measured again then and measure territory signal y
b
Step 7. will be measured territory signal y
bBe transferred to reconstructed module, utilize the compressed sensing restructing algorithm that current video block is reconstructed.
The correlation of the sub-piece of the plain expression of cross covariance entry of a matrix and the variance of sub-piece itself, therefore compared with prior art, the measurement territory block sort method that the present invention proposes has not only been considered the global statistics characteristic of video block signal, has also considered the local distribution characteristic of video block signal to a certain extent.
It is as follows that realization of the present invention is that also step 3 is measured the establishment step of correlative relationship model of territory signal and frequency-region signal:
1) according to the character of measuring matrix, set up the relational model of measuring territory covariance matrix and frequency-region signal:
Wherein, C
yFor measuring the cross covariance matrix of characteristic of field image, m is for measuring number, and θ is DCT territory sample matrix;
2) according to the characteristic distributions of vision signal in the DCT territory, set up the relational model of frequency domain cross covariance matrix and frequency-region signal:
Wherein, C
θBe the cross covariance matrix on the DCT territory, n is the pixel count of video piece, and θ is DCT territory sample matrix;
3) by step 1) and 2) relational model set up to measure the correlative relationship model of territory signal and frequency-region signal:
Wherein, C
yFor measuring the cross covariance matrix of characteristic of field image, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory.
The present invention has designed based on the video compression perceptual coding system of measuring the territory block sort, provides based on the video compression perceptual coding method of measuring the territory block sort, compared with prior art has following advantage:
(1) the present invention has realized measuring the video block sort on the territory, under coding side pixel domain signal non-availability or the condition that is not reconstructed, still can realize the classification of video piece, carry out in various degree compression according to the characteristics of dissimilar video pieces, improved the distortion performance of system;
(2) the present invention has set up the correlative relationship model of measuring territory signal and frequency-region signal, utilizes and measures the correlation analysis that the territory signal has been realized frequency-region signal;
(3) the block sort method in the measurement territory of the present invention's realization has not only been considered the statistical property of current video block, and has considered the local distribution characteristic of video piece to a certain extent, has improved the accuracy of block sort.
Description of drawings
Fig. 1 is the formation block diagram that the present invention is based on the video compression perceptual coding system of measuring the territory block sort.
Fig. 2 is method flow schematic diagram of the present invention.
Fig. 3 is the experimental result of the measurement territory block sort of lena image of the present invention, and wherein Fig. 3 (a) is original image, and Fig. 3 (b), 3 (c) and 3 (d) are respectively Fig. 3 (a) through smooth block, edge block and texture block after the block sort.
Embodiment
Be described in further detail below in conjunction with accompanying drawing 1 and the present invention of Fig. 2:
Step 1. piecemeal compression sampling
With input picture piece or video piece x
bThe sub-piece that is divided into the equal and opposite in direction non-overlapping copies is measured matrix to each sub-piece with the gaussian random of identical size and is measured measured value respectively.Specific embodiments of the invention are that the video piece that will intend sampling is further divided into four sub-piece x
B1, x
B2, x
B3, x
B4, corresponding DCT territory signal is θ
1, θ
2, θ
3, θ
4, each sub-piece to be measured matrix with the gaussian random of identical size sample respectively, sample rate is MR=0.5, obtains measuring territory signal y respectively
1, y
2, y
3, y
4, measuring the territory signal is the column vector of m * 1;
Step 2. is extracted and is measured the characteristic of field image
The measurement territory signal of sub-piece is measured the characteristic of field image as the row vector structure of matrix
Step 3. is asked the cross covariance matrix of measuring the characteristic of field image
3a) set up the correlative relationship model of measuring territory signal and frequency-region signal:
Wherein, C
yFor measuring the cross covariance matrix of characteristic of field image, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory;
3b) ask the cross covariance matrix of measuring the characteristic of field image, specific embodiments of the invention obtain:
Wherein, matrix element C
Ij=Cov (y
i, y
j)=E{[y
i-E (y
i)] [y
j-E (y
j)] be second order mixing center square, 1≤i≤4,1≤j≤4.
Diagonal entry C
Ij(i=j)=E{[y
i-E (y
i)]
2Expression measurement territory signal y
iVariance, off diagonal element C
Ij(the sub-piece y of the expression of i ≠ j)
iAnd y
jBetween correlation.
Step 4. pair is measured each element of territory cross covariance matrix and is asked variance, by step 3a) relational model get:
Wherein, variance, C are asked in var () expression
yFor measuring the cross covariance matrix of characteristic of field image, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory.
Step 5. is adjudicated the type of video piece by following criterion, is judged to smooth block, edge block or texture block according to the big young pathbreaker's video of the variance piece of the cross covariance matrix element of measuring the characteristic of field image:
Wherein if be " if " the meaning, T
1, T
2Be empirical value.
Get T in the specific embodiments of the invention
1=3 * 10
6, T
2=2 * 10
7
Step 6. sample rate is distributed and secondary is measured
Video block type according to judging module output carries out the sample rate distribution, edge block＞texture block＞smooth block is satisfied in the distribution of sample rate, determine hits and measure matrix according to the size of sample rate and video piece, utilize this measurement matrix that current video block is measured again then and measure the territory signal.In the specific embodiments of the invention video piece is divided into edge block, texture block and smooth block, each video piece sample rate is taken as 0.5,0.4 and 0.3 respectively.
Step 7. will be measured the territory signal and be transferred to reconstructed module, utilize the compressed sensing restructing algorithm that current video block is reconstructed.Adopt base to follow the trail of (Basic Pursuit in the specific embodiments of the invention, BP) algorithm is reconstructed, adopt the dictionary of adjacent reconstructing video frame structure (to see document Liu Haixiao as sparse base, Song Bin, Qin Hao, Qiu Zhiliang, A Dictionary Generation Scheme for Block-Based Compressed Video Sensing, Proceedings of ICSPCC, Sep 2011).
Be described further in the block sort effect of measuring on the territory below in conjunction with 3 couples of the present invention of accompanying drawing:
Realize the simulated conditions of Fig. 3: hardware environment: CPU AMD Sempron3000+, 1.8GHz, 512MB internal memory; Software environment: Windows XP, Matlab R2010b; Image block size: 8 * 8; Reference picture: lena; Resolution: 256 * 256.
As can be seen from Figure 3, the measurement territory block sort method that the present invention proposes can be carried out block sort to image or frame of video preferably, can more accurately image 3 (a) be divided into smooth block 3 (b), edge block 3 (c) and texture block 3 (d) basically.
Couple the present invention is described further in the effect aspect the objective quality below in conjunction with table 1:
Table 1 be the present invention and prior art based on the Y-PSNR of the method for video coding of compressed sensing (Peak Signal to Noise Ratio, PSNR) and the comparing result of scramble time.Realize the simulated conditions of this table: hardware environment: CPU AMD Sempron3000+, 1.8GHz, 512MB internal memory; Software environment: Windows XP, Matlab R2010b; Video piece size: 8 * 8; Reference sequences: Carphone, News, Hall_monitor, Mother_daughter, Stefan; The frame number of cycle tests: first frame of all sequences; Resolution: 176 * 144 (Carphone, News), 352 * 288 (Hall_monitor, Mother_daughter, Stefan); Measure matrix: gaussian random is measured matrix; Restructing algorithm: basic tracing algorithm; Sparse base: the dictionary of adjacent reconstructing video frame structure.
As can be seen from Table 1, the present invention compares based on the video coding technique of compressed sensing with existing, and the Y-PSNR PSNR of its decoding and reconstituting frame of video has improved about 0.3～1.3dB.Because the encoding and decoding time that the introducing of block sort method herein increases is no more than 5% of the whole system encoding and decoding time, does not increase the complexity of system basically.Equate owing to two kinds of used overall measurements of method frame of video count up to entirely again, so the present invention has improved the distortion performance of whole system.
Table 1
Parameter in above-mentioned Fig. 3 and the form has further been verified video compression perceptual coding system and the method based on measurement territory block sort that the present invention proposes, compare based on the method for video coding of compressed sensing with prior art, improved the quality of reconstructing video and the distortion performance of whole video coded system.
Claims (4)
1. the video compression perceptual coding system based on measurement territory block sort comprises that piecemeal compression sampling module, the block sort module of measuring the territory, sample rate distribution module, hits distribute and secondary measurement module and reconstructed module; Piecemeal compression sampling module is divided into the sub-piece of equal and opposite in direction non-overlapping copies with the input video piece, each sub-piece is measured with identical random measurement matrix measure the territory signal; Measure the block sort module in territory and measure the characteristic of field image module by extracting, measuring territory and frequency domain relational model module and judging module constitutes, the measurement territory signal of the sub-piece that the piecemeal compression sampling is obtained obtains measuring the characteristic of field image as the capable vector of matrix respectively, measure territory and frequency domain relational model module and set up the linear approximate relationship of measuring the correlation of territory signal and corresponding frequency-region signal, judging module is that theoretical foundation utilization measurement characteristic of field image carries out the judgement of piece classification with the correlative relationship model of measuring territory and frequency-region signal, be specially and calculate the cross covariance matrix of measuring the characteristic of field image, and drawn the variance of cross covariance entry of a matrix element by the correlative relationship model for the feature of weighing the video piece, realize the type judgement of video piece according to the size of variance yields, by judging module the video piece of input is judged to edge block, texture block or smooth block; System distributes different sample rates at the characteristics of court verdict and dissimilar video piece itself to dissimilar video pieces then; Hits distributes and the secondary measurement module is determined to measure number and measured matrix according to sample rate and video block sizes, utilizes to measure matrix and the video piece is measured again measure the territory signal, will measure the territory signal at last and be transferred to reconstructed module and be reconstructed.
2. according to claim 1 based on the video compression perceptual coding system of measuring the territory block sort, it is characterized in that: described foundation measurement territory and frequency domain relational model are weighed the correlation of two signals with cross covariance, through the coefficient characteristic distributions after the discrete cosine transform, set up the linear approximate relationship model of measuring between territory signal and the corresponding frequency domain signal correlation according to vision signal.
3. one kind based on the video compression perceptual coding method of measuring the territory block sort, and the system that is suitable for is characterized in that with claim 1: this method comprises:
Step 1. is with input picture piece or video piece x
bThe sub-piece that is divided into the equal and opposite in direction non-overlapping copies supposes to be divided into four sub-piece x
B1, x
B2, x
B3, x
B4, corresponding DCT territory signal is θ
1, θ
2, θ
3, θ
4, each sub-piece is measured matrix with the gaussian random of identical size measures measurement territory signal y respectively
1, y
2, y
3, y
4, measuring the territory signal is the column vector of m * 1;
Step 2. is measured characteristic of field image y=[y with the measurement territory signal of sub-piece as the row vector structure of matrix
1, y
2, y
3, y
4]
T
Step 3. is set up the correlative relationship model of measuring territory signal and frequency-region signal:
Wherein, C
yFor measuring the cross covariance matrix of characteristic of field image, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory;
Step 4. pair is measured each element of the cross covariance matrix of characteristic of field image and is asked variance, is got by the relational model of step 3:
Wherein, variance, C are asked in var () expression
yFor measuring the cross covariance matrix of characteristic of field image, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory;
Step 5. is adjudicated the type of video piece by following criterion, is judged to smooth block, edge block or texture block according to the big young pathbreaker's video of the variance piece of the cross covariance matrix element of measuring the characteristic of field image:
Wherein, if be " if " the meaning, variance, C are asked in var () expression
yFor measuring the cross covariance matrix of characteristic of field image, T
1, T
2Be empirical value;
The video piece of step 6. pair judgement carries out sample rate and distributes, edge block＞texture block＞smooth block is satisfied in the distribution of sample rate, determine hits and measure matrix according to the size of sample rate and video piece, utilize this measurement matrix that current video block is measured again then and measure territory signal y
b
Step 7. will be measured territory signal y
bBe transferred to reconstructed module, utilize the compressed sensing restructing algorithm that current video block is reconstructed.
4. according to claim 3 based on the video compression perceptual coding method of measuring the territory block sort, it is characterized in that: the establishment step of the correlative relationship model of described step 3 measurement territory signal and frequency-region signal is as follows:
4.1) according to the character of measuring matrix, set up the relational model of measuring territory cross covariance matrix and frequency-region signal:
Wherein, C
yFor measuring the cross covariance matrix of characteristic of field image, m is for measuring number, and θ is DCT territory sample matrix;
4.2) according to the characteristic distributions of vision signal in the DCT territory, set up the relational model of frequency domain covariance matrix and frequency-region signal:
Wherein, C
θBe the cross covariance matrix on the DCT territory, n is the pixel count of video piece, and θ is DCT territory sample matrix;
4.3) by step 4.1) and 4.2) relational model set up to measure the correlative relationship model of territory signal and corresponding frequency-region signal:
Wherein, C
yFor measuring the cross covariance matrix of characteristic of field image, n is the pixel count of video piece, and m is for measuring number, C
θBe the cross covariance matrix on the DCT territory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310089235.8A CN103338363B (en) | 2013-03-06 | 2013-03-06 | Video compress perceptual coding system based on measurement field block sort and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310089235.8A CN103338363B (en) | 2013-03-06 | 2013-03-06 | Video compress perceptual coding system based on measurement field block sort and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103338363A true CN103338363A (en) | 2013-10-02 |
| CN103338363B CN103338363B (en) | 2016-08-03 |
Family
ID=49246453
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310089235.8A Expired - Fee Related CN103338363B (en) | 2013-03-06 | 2013-03-06 | Video compress perceptual coding system based on measurement field block sort and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103338363B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015076634A1 (en) * | 2013-11-24 | 2015-05-28 | Lg Electronics Inc. | Method and apparatus for encoding and decoding video signal using adaptive sampling |
| CN105049870A (en) * | 2015-08-05 | 2015-11-11 | 南京邮电大学 | Sparseness estimation-based distributed video compressed sensing fast reconstruction method |
| CN105306936A (en) * | 2015-07-17 | 2016-02-03 | 福州大学 | BCS (Block Compressive Sensing)-based image coding method |
| CN111986280A (en) * | 2020-07-22 | 2020-11-24 | 广西交通设计集团有限公司 | Image compression method based on maximum inter-class variance block compressed sensing |
| CN113055678A (en) * | 2021-03-06 | 2021-06-29 | 复旦大学 | Measurement domain compressed sensing coding algorithm based on adjacent pixel correlation |
| CN115393777A (en) * | 2022-10-28 | 2022-11-25 | 国网山东省电力公司青岛供电公司 | Electric power video monitoring image edge calculation method and system based on compressed sensing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722896A (en) * | 2012-05-22 | 2012-10-10 | 西安电子科技大学 | Adaptive compressed sensing-based non-local reconstruction method for natural image |
| US20120286981A1 (en) * | 2011-05-12 | 2012-11-15 | Texas Instruments Incorporated | Compressive sensing analog-to-digital converters |
-
2013
- 2013-03-06 CN CN201310089235.8A patent/CN103338363B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120286981A1 (en) * | 2011-05-12 | 2012-11-15 | Texas Instruments Incorporated | Compressive sensing analog-to-digital converters |
| CN102722896A (en) * | 2012-05-22 | 2012-10-10 | 西安电子科技大学 | Adaptive compressed sensing-based non-local reconstruction method for natural image |
Non-Patent Citations (3)
| Title |
|---|
| MASOMEH AZGHANI等: "Compressed Video Sensing Using Adaptive Sampling Rate", 《2010 5TH INTERNATIONAL SYMPOSIUM ON TELECOMMUNICATIONS (IST"2010)》 * |
| ZHAORUI LIU等: "BLOCK-BASED ADAPTIVE COMPRESSED SENSING FOR VIDEO", 《PROCEEDINGS OF 2010 IEEE 17TH INTERNATIONAL CONFERENCE ON IMAGE PROCESSING》 * |
| 张淑芳等: "基于压缩感知的图像自适应编码算法", 《天津大学学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015076634A1 (en) * | 2013-11-24 | 2015-05-28 | Lg Electronics Inc. | Method and apparatus for encoding and decoding video signal using adaptive sampling |
| US10462464B2 (en) | 2013-11-24 | 2019-10-29 | Lg Electronics Inc. | Method and apparatus for encoding and decoding video signal using adaptive sampling |
| CN105306936A (en) * | 2015-07-17 | 2016-02-03 | 福州大学 | BCS (Block Compressive Sensing)-based image coding method |
| CN105306936B (en) * | 2015-07-17 | 2018-06-15 | 福州大学 | A kind of method for encoding images based on block compressed sensing |
| CN105049870A (en) * | 2015-08-05 | 2015-11-11 | 南京邮电大学 | Sparseness estimation-based distributed video compressed sensing fast reconstruction method |
| CN105049870B (en) * | 2015-08-05 | 2018-05-01 | 南京邮电大学 | A kind of distributed video compressed sensing fast reconstructing method based on degree of rarefication estimation |
| CN111986280A (en) * | 2020-07-22 | 2020-11-24 | 广西交通设计集团有限公司 | Image compression method based on maximum inter-class variance block compressed sensing |
| CN111986280B (en) * | 2020-07-22 | 2021-06-01 | 广西交通设计集团有限公司 | Image compression method based on maximum inter-class variance block compressed sensing |
| CN113055678A (en) * | 2021-03-06 | 2021-06-29 | 复旦大学 | Measurement domain compressed sensing coding algorithm based on adjacent pixel correlation |
| CN113055678B (en) * | 2021-03-06 | 2022-05-20 | 复旦大学 | Measurement domain compressed sensing coding algorithm based on adjacent pixel correlation |
| CN115393777A (en) * | 2022-10-28 | 2022-11-25 | 国网山东省电力公司青岛供电公司 | Electric power video monitoring image edge calculation method and system based on compressed sensing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103338363B (en) | 2016-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103338363A (en) | System and method for video compression sensing and encoding based on measurement domain block classification | |
| CN100559881C (en) | A kind of method for evaluating video quality based on artificial neural net | |
| CN103037212B (en) | The adaptive block compressed sensing method for encoding images of view-based access control model perception | |
| CN101715146B (en) | Method and system for evaluating quality of compressed video | |
| CN105761290B (en) | A kind of adaptive multiple dimensioned splits' positions perception method of sampling | |
| CN102982805B (en) | Multi-channel audio signal compressing method based on tensor decomposition | |
| CN101478691B (en) | Non-reference evaluation method for Motion Jpeg2000 video objective quality | |
| Wang et al. | Novel spatio-temporal structural information based video quality metric | |
| CN102685548B (en) | The nothing ginseng appraisal procedure of video quality | |
| CN105872555B (en) | A kind of steganalysis algorithm for the insertion of H.264 video motion vector information | |
| CN106941609B (en) | Video-frequency compression method based on the perception of adaptive splits' positions | |
| CN102427527A (en) | Method for reconstructing non key frame on basis of distributed video compression sensing system | |
| CN101742353A (en) | No-reference video quality evaluating method | |
| CN102123278A (en) | Signal source encoding method based on distributed compressive sensing technology | |
| CN102148987A (en) | Compressed sensing image reconstructing method based on prior model and 10 norms | |
| CN103200421A (en) | No-reference image quality evaluation method based on Curvelet transformation and phase coincidence | |
| CN102387365B (en) | Adaptive image coding method based on compressive sensing | |
| CN103440675A (en) | Overall situation reconstitution optimization model construction method for image block compressed sensing | |
| CN103237204A (en) | Video signal collection and reconfiguration system based on high-dimension compressed sensing | |
| CN105678699B (en) | Compressed sensing image reconstructing method based on the detection of measurement field piecemeal conspicuousness | |
| CN102946539B (en) | Method for estimating motion among video image frames based on compressive sensing | |
| Song et al. | Compressed sensing image reconstruction using intra prediction | |
| CN103354617A (en) | Boundary strength compressing image quality objective evaluation method based on DCT domain | |
| CN103686178A (en) | Method for extracting area-of-interest of video based on HVS | |
| CN103714534A (en) | Material surface defect detection method based on compressed sensing |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160803 Termination date: 20210306 |