CN101668211B - Method for encoding and decoding images and videos - Google Patents
Method for encoding and decoding images and videos Download PDFInfo
- Publication number
- CN101668211B CN101668211B CN 200910152902 CN200910152902A CN101668211B CN 101668211 B CN101668211 B CN 101668211B CN 200910152902 CN200910152902 CN 200910152902 CN 200910152902 A CN200910152902 A CN 200910152902A CN 101668211 B CN101668211 B CN 101668211B
- Authority
- CN
- China
- Prior art keywords
- image
- image object
- mathematical modeling
- concrete
- decoder
- 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 22
- 238000007906 compression Methods 0.000 claims abstract description 11
- 230000006835 compression Effects 0.000 claims abstract description 11
- 239000000463 material Substances 0.000 claims description 6
- 230000000295 complement effect Effects 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims 1
- 230000008676 import Effects 0.000 claims 1
- 230000005540 biological transmission Effects 0.000 abstract description 10
- 230000000694 effects Effects 0.000 abstract description 5
- 230000005477 standard model Effects 0.000 abstract 3
- 238000013144 data compression Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000005056 compaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000008921 facial expression Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
Images
Abstract
The invention relates to a method for encoding and decoding images and videos. The current method influences the quality of videos, increases the risk of data loss simultaneously and has poor encoding compression rate. The method in the invention comprises the following steps: firstly classifying single image objects; building image object standard model databases; respectively implanting identical standard model databases into an image encoder and an image decoder; then carrying out encoding on the images by utilizing the image encoder; retrieving the standard model databases and rebuilding the image objects by the image decoder; combining the rebuilt image objects into virtual images; and obtaining the real images by combing with the virtual images according to image residue error. The method in the invention greatly reduces the data quantity used for channel transmission, and when the quantity of the images needing to be processed is enormous, the effect is more superior and image storage and retrieval are more convenient.
Description
Technical field
The invention belongs to image and video signal treatment technique field, be specifically related to the method for a kind of image and coding and decoding video.
Background technology
Visual activity is one of human most important basic activity, and visual information is main path and the important form that the mankind obtain external knowledge, the understanding world.According to statistics, human nearly information more than 80% is obtained by vision.Image Information Processing is generation and the collection to picture signal of research people or machine, and the formation of information, extraction, analysis, comprehensive, the theory and the science of method expressing and utilize are one and have concurrently comprehensive and subject intercrossing.Image and vision signal are handled and are mainly comprised: image and video data compaction coding, transmission, decoding, storage, retrieval etc., and compressed encoding and decoding are prerequisite and basis.
In international multimedia technology field, the rest image JPEG2000 standard that released one after another, video MPEG-4 and H.264 wait multi-medium data compressed encoding international standard.Wherein, object-oriented and content-based video coding and decoding technology extract coding respectively with the content that showed especially dynamic content from video, saved the mass data amount, and the degree difference of being concerned about according to contents of object, to its reinforcement or painstakingly do not strengthen descriptor, i.e. time-domain extended coding and spatial domain extended coding.The time-domain extended coding is to increase constantly in key operations to insert some intermediate frames, and it is clear to make its action seem to link up more; The spatial domain extended coding is to giving careful description such as position extended resolutions such as facial expressions.Like this, content-based coding and decoding video makes the data compression effect enlarge several times than other compression methods, promptly compresses the back data transfer rate still less, but much better to the expression effect of image.
Yet there are a lot of problems in existing video and image encoding and decoding technique:
(1) the data volume difference between video I frame and P frame or the B frame is very big, makes transmission code rate peak value and rate of change big ups and downs, influences transmission quality, and need take more transmission bandwidth;
(2) I frame data amount is very big, and compression effectiveness is not good, and for improving compression ratio, the I frame need be provided with more sparsely, in case but so I LOF can make following closely P frame or B frame decode again, influence video quality, increased the risk of loss of data;
(3) the still image coding compression ratio can't be satisfactory, and can object-oriented and content-based encoding and decoding technique be applied to still image coding is a problem that waits to explore.
Summary of the invention
The objective of the invention is at the deficiencies in the prior art, the method for a kind of image and coding and decoding video is provided, this method is applicable to rest image or video I two field picture.
The inventive method comprises the steps:
Step (1) is set up image object master pattern database, and concrete grammar is:
A, set up the Still image data storehouse, large nuber of images is analyzed, each two field picture is divided into single image object according to picture material;
B, single image object is classified according to different things classifications, such as people, automobile, animal and building;
C, the image object of each type is set up Mathematical Modeling, this model uses semanteme or mathematical linguistics to describe, and the image of the type can be rebuild or duplicate fully according to this semanteme or Mathematical Modeling;
D, all types of Mathematical Modelings are created as the master pattern database, and identical master pattern database is implanted image encoder and image decoder respectively;
Step (2) utilizes image encoder that image is encoded, and the concrete steps of coding are:
E, input one frame image to be encoded;
F, image encoder are divided into image object according to picture material;
G, according to image object automatic Mathematical Modeling of being complementary of retrieval in the master pattern database;
H, give Mathematical Modeling with real parameter, obtain the Mathematical Modeling of the concrete parametric description of usefulness of corresponding each image object according to the objective attribute of image object;
I, the concrete parameter of basis are rebuild virtual image object and are combined into virtual image;
J, with virtual image and true picture relatively carries out estimation and compensation, obtains the image residual error of virtual image with respect to true picture;
K, concrete parameter, the image object of Mathematical Modeling is combined into combination parameter that virtual image generates and image residual error constitutes compress coding data and be transferred to channel;
Step (3) is decoded to the image compression encoding data of image decoder to channel, and the concrete steps of decoding are:
L, decoder are according to the concrete parameter of Mathematical Modeling, search criteria model database reconstructed image object;
M, decoder are combined into virtual image according to combination parameter with the image object of rebuilding;
N, decoder are according to the image residual error, and the combined with virtual image obtains true picture.
The present invention is a cost to increase image compression encoding and decoding complexity, greatly reduce the data volume that is used for Channel Transmission, this technical thought meets such objective reality: promptly the bandwidth resources than channel are more abundant far away for the calculating at information source and stay of two nights two ends and storage resources.Make it suitable with the B frame by the data volume that reduces video I frame, further improve data compression ratio, transmission code rate peak value and rate of change are fluctuateed gently, take the transmission bandwidth resource and greatly reduce with the P frame.Also can adopt content-based compression coding and decoding algorithm to rest image, effect is more superior when the amount of images of needs processing is huge, also be more convenient for image storage and retrieval.
Description of drawings
Fig. 1 is a flow chart of the present invention.
Embodiment
As shown in Figure 1, the concrete steps of the inventive method are:
Step (1) is set up image object master pattern database, and concrete grammar is:
A, set up the Still image data storehouse, large nuber of images is analyzed, each two field picture is divided into single image object according to picture material;
B, single image object is classified according to different things classifications, such as people, car, dog and house;
C, the image object of each type is set up Mathematical Modeling, this model uses semanteme or mathematical linguistics to describe, and the image of the type can be rebuild or duplicate fully according to this semanteme or Mathematical Modeling;
D, all types of Mathematical Modelings are created as the master pattern database, and identical master pattern database is implanted image encoder and image decoder respectively;
Step a～steps d has been described the method for setting up image object master pattern database, is the preliminary preparation and the basis of Image Data Compression code decode algorithm;
Step (2) utilizes image encoder that image is encoded, and the concrete steps of coding are:
E, input one frame image to be encoded;
F, image encoder are divided into image object according to picture material;
G, according to image object automatic Mathematical Modeling of being complementary of retrieval in the master pattern database;
H, give Mathematical Modeling with real parameter, obtain the Mathematical Modeling of the concrete parametric description of usefulness of corresponding each image object according to the objective attribute of image object;
I, the concrete parameter of basis are rebuild virtual image object and are combined into virtual image;
J, with virtual image and true picture relatively carries out estimation and compensation, obtains the image residual error of virtual image with respect to true picture;
K, concrete parameter, the image object of Mathematical Modeling is combined into combination parameter that virtual image generates and image residual error constitutes compress coding data and be transferred to channel;
Step e～step k has described the cataloged procedure of a two field picture.
Step (3) is decoded to the image compression encoding data of image decoder to channel, and the concrete steps of decoding are:
L, decoder are according to the concrete parameter of Mathematical Modeling, search criteria model database reconstructed image object;
M, decoder are combined into virtual image according to combination parameter with the image object of rebuilding;
N, decoder are according to the image residual error, and the combined with virtual image obtains true picture.
Step 1～step n describes the decode procedure of a two field picture.
The embodiment of the invention is described the compression coding and decoding method of a frame rest image or a frame video image (referring in particular to the I frame), P frame or B frame for video image, the estimation before then continuing and the technical scheme of compensation, like this, data volume sizableness between I frame and P frame or the B frame, further improve data compression ratio, transmission code rate peak value and rate of change are fluctuateed gently, take the transmission bandwidth resource and greatly reduce.Simultaneously, the I frame can be provided with very closely, reduces the risk of loss of data.
Claims (1)
1. the method for image and coding and decoding video is characterized in that this method comprises the steps:
Step (1). set up image object master pattern database, concrete grammar is:
A. set up the Still image data storehouse, large nuber of images is analyzed, each two field picture is divided into single image object according to picture material;
B. single image object is classified according to different things classifications;
C. the image object of each type is set up Mathematical Modeling, this model uses semanteme or mathematical linguistics to describe, and the image of the type can be rebuild or duplicate fully according to this semanteme or Mathematical Modeling;
D. all types of Mathematical Modelings are created as the master pattern database, and identical master pattern database is implanted image encoder and image decoder respectively;
Step (2). utilize image encoder that image is encoded, concrete steps are:
E. import frame image to be encoded;
F. image encoder becomes image object according to the picture material image segmentation that one frame is to be encoded;
G. in the master pattern database, retrieve the Mathematical Modeling that is complementary automatically according to image object;
H. give Mathematical Modeling with real parameter according to the objective attribute of image object, obtain the Mathematical Modeling of the concrete parametric description of usefulness of corresponding each image object;
I. rebuild virtual image object and be combined into virtual image according to concrete parameter;
J. virtual image and true picture are compared, carry out estimation and compensation, obtain the image residual error of virtual image with respect to true picture;
K. concrete parameter, the image object of Mathematical Modeling being combined into combination parameter that virtual image generates and image residual error constitutes compress coding data and is transferred to channel;
Step (3). channel is decoded to the image compression encoding data of image decoder, and concrete steps are:
L. image decoder is according to the concrete parameter of Mathematical Modeling, search criteria model database reconstructed image object;
M. image decoder is combined into virtual image according to combination parameter with the image object of rebuilding;
N. image decoder is according to the image residual error, and the combined with virtual image obtains true picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910152902 CN101668211B (en) | 2009-09-18 | 2009-09-18 | Method for encoding and decoding images and videos |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910152902 CN101668211B (en) | 2009-09-18 | 2009-09-18 | Method for encoding and decoding images and videos |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101668211A CN101668211A (en) | 2010-03-10 |
| CN101668211B true CN101668211B (en) | 2011-08-10 |
Family
ID=41804612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910152902 Active CN101668211B (en) | 2009-09-18 | 2009-09-18 | Method for encoding and decoding images and videos |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101668211B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103973634B (en) * | 2013-01-24 | 2015-03-18 | 腾讯科技（深圳）有限公司 | Application data construction method, related equipment and network system |
| CN105578192A (en) * | 2015-12-16 | 2016-05-11 | 国网浙江省电力公司湖州供电公司 | Power visual metamodel agglomeration compression method |
| CN107018421B (en) * | 2016-01-27 | 2019-08-23 | 北京中科晶上科技有限公司 | A kind of image sending, receiving method and device, system |
| CN109547786B (en) * | 2017-09-22 | 2023-05-09 | 阿里巴巴集团控股有限公司 | Video encoding and video decoding methods and devices |
| CN108924572A (en) * | 2018-07-26 | 2018-11-30 | 高新兴科技集团股份有限公司 | A kind of computer graphical and the adaptive video coding-decoding method and system of real scene image |
-
2009
- 2009-09-18 CN CN 200910152902 patent/CN101668211B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN101668211A (en) | 2010-03-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101668211B (en) | Method for encoding and decoding images and videos | |
| WO2021208247A1 (en) | Mimic compression method and apparatus for video image, and storage medium and terminal | |
| CN106485656B (en) | A kind of method of image super-resolution reconstruct | |
| CN110517329A (en) | A kind of deep learning method for compressing image based on semantic analysis | |
| CN105323503B (en) | A kind of panoramic video transmission method and system | |
| CN104539961B (en) | Gradable video encoding system based on the gradual dictionary learning of hierarchy | |
| CN110166779A (en) | Video-frequency compression method based on super-resolution reconstruction | |
| WO2016095696A1 (en) | Video-outline-based method for monitoring scalable coding of video | |
| WO2022121649A1 (en) | Point cloud data encoding and decoding method, point cloud data processing method and apparatus, electronic device, computer program product, and computer readable storage medium | |
| CN113192147B (en) | Method, system, storage medium, computer device and application for significance compression | |
| CN107454412A (en) | A kind of processing method of video image, apparatus and system | |
| Ma | Distributed video coding scheme of multimedia data compression algorithm for wireless sensor networks | |
| CN110460882B (en) | Stream media content distribution method based on data feature dimension reduction coding | |
| Xu et al. | Singular vector sparse reconstruction for image compression | |
| CN105590296B (en) | A kind of single-frame images Super-Resolution method based on doubledictionary study | |
| CN107292821B (en) | A kind of super-resolution image reconstruction method and system | |
| Chattopadhyay et al. | Human motion capture data compression by model-based indexing: A power aware approach | |
| CN103533353B (en) | A kind of near video coding system | |
| WO2023184980A1 (en) | Codebook-based method for performing image compression using self-encoding machine | |
| CN103561264A (en) | Media decoding method based on cloud computing and decoder | |
| CN116416216A (en) | Quality evaluation method based on self-supervision feature extraction, storage medium and terminal | |
| CN107018421B (en) | A kind of image sending, receiving method and device, system | |
| CN115131254A (en) | Constant bit rate compressed video quality enhancement method based on two-domain learning | |
| CN113132755B (en) | Method and system for encoding extensible man-machine cooperative image and method for training decoder | |
| CN104519360B (en) | A kind of compression method based on HEVC |
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 | ||
| ASS | Succession or assignment of patent right |
Owner name: SHANGHAI BLOZI ELECTRIC CO., LTD. Free format text: FORMER OWNER: HANGZHOU ELECTRONIC SCIENCE AND TECHNOLOGY UNIV Effective date: 20150122 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 310018 TO: 200433 |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20150122 Address after: 200433, No. 5, building 103, Fudan Science Park, 1018 Miyun Road, Shanghai, Yangpu District Patentee after: SHANGHAI BLOZI ELECTRONIC CO.,LTD. Address before: Hangzhou City, Zhejiang province 310018 Jianggan District Xiasha Higher Education Park No. 2 street Patentee before: HANGZHOU DIANZI University |
|
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 201899 financial Valley 23B, No. 1818, North Road, Jiading District, Shanghai Patentee after: SHANGHAI BAOZI INTELLIGENT TECHNOLOGY CO.,LTD. Address before: 200433, No. 5, building 103, Fudan Science Park, 1018 Miyun Road, Shanghai, Yangpu District Patentee before: SHANGHAI BLOZI ELECTRONIC CO.,LTD. |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20240129 Address after: No. 288 Yueling Road, Lishan District, Anshan City, Liaoning Province, 114044 Patentee after: Qixin optoelectronics Co.,Ltd. Country or region after: China Address before: 23B, financial Valley, 1818 Chengbei Road, Jiading District, Shanghai, 201899 Patentee before: SHANGHAI BAOZI INTELLIGENT TECHNOLOGY CO.,LTD. Country or region before: China |
|
| TR01 | Transfer of patent right |