CN100435550C - Method for increasing anti-bit-error ability of grade-tree collecting splitting algorithm coding-decoding device - Google Patents
Method for increasing anti-bit-error ability of grade-tree collecting splitting algorithm coding-decoding device Download PDFInfo
- Publication number
- CN100435550C CN100435550C CNB2006100118672A CN200610011867A CN100435550C CN 100435550 C CN100435550 C CN 100435550C CN B2006100118672 A CNB2006100118672 A CN B2006100118672A CN 200610011867 A CN200610011867 A CN 200610011867A CN 100435550 C CN100435550 C CN 100435550C
- Authority
- CN
- China
- Prior art keywords
- coding
- mistake
- importance
- bag
- scanning
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The present invention provides a method for increasing error code resisting capability of grade-tree collecting splitting algorithm encoders and decoders, which comprises an encoding method and a corresponding decoding method. The present invention is characterized in that the encoding method comprises the following steps: wavelet transformation is carried out for input images; a lowest-frequency region obtained after the wavelet transformation is divided to form a plurality of n*n matrixes; afterwards, tree collection in a corresponding space direction in each matrix is encoded and the encoding result of each matrix is packed. The method can effectively control the large-scale spread of the error of error codes and enhance the fault freedom of the grade-tree collecting splitting algorithm (SPIHT).
Description
Technical field
The present invention relates to the rest image encoding and decoding technique, particularly a kind of method of the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades, adopt this method can effectively control the propagation on a large scale of error code mistake, strengthen the fault freedom of hierarchical tree set partitioning algorithm (SPIHT).
Background technology
In the current information age, network is ubiquitous, and the storage of image and transmission then become more and more important.And will be at limited capacity, and have and propagate the image that comprises huge amount of information in the channel of noise, just essential to the compression of view data so.So the research to image compression encoding is more and more, wherein particularly active with the lossy compression method field.Though these methods are improved constantly code efficiency, Bian Ma complexity also improves thereupon simultaneously.Yet the appearance of zero-tree wavelet coding (EZW) has changed this situation, and this algorithm has many excellent characteristic,, embedded bitstream low as complexity, controllable compression ratios rate or the like.Because the EZW coding method has many advantages, so a lot of expert both at home and abroad begins one's study and makes improvements, have the greatest impact with hierarchical tree set partitioning algorithm (SPIHT), in the stationary image compression coding field one of the highest algorithm of the code efficiency of generally acknowledging.
Spiht algorithm has defined a kind of set partitioning rule: 1. O (i, j): node (i, the set of the coordinate of the direct child node of j) all (Offspring); 2. D (i, j): node (i, the set of the coordinate of all child nodes (Descendants) j); 3. H: after the wavelet decomposition top in the set (comprising high and low frequency) of coordinate of all nodes; 4. L (i j) is the set of all non-direct offsprings' node coordinate, L (i, j)=D (i, j)-O (i, j).In the specific implementation of spiht algorithm, conspicuousness information is kept in three tables of arranging preface, they are respectively not significantly set table (LIS), significantly pixel table (LIP) and significantly pixel table (LSP).Just because of this set splicing mechanism, just be achieved its high efficiency coding.
SPIHT coding method implementation method comprises following step as shown in Figure 1:
(1) input picture is done wavelet transformation and obtain wavelet coefficient matrix P;
(2) the maximum max of coefficient among the compute matrix P, and generate a matrix Q, dimension is 1/2 of P, and (i j) is corresponding D (i, j) maximum in to element wherein.
(3) initialization iterations n and threshold value T, inessential coefficient table (LIP), significant coefficient table (LSP), inessential set table (LIS);
(4) to LIP, the element in the LIS table carries out importance scanning (Sorting Pass);
(5) element among the LSP is carried out refinement scanning (Refinement Pass);
(6) repeating (3) (4), is 0 up to n, and perhaps compression ratio reaches requirement;
(7) compressed code flow is carried out adaptive arithmetic code;
Its core encoder partly is (3) (4) (5), and implementation method below is the course of work of core encoder as shown in Figure 2 in detail:
At first, calculate n=|log
2(max) |, T=2
n, and will own (i, j) coordinate of ∈ H is added among the LIP, the low frequency coefficient that is about among the H joins among the LIP, simultaneously with all child nodes (i, j) coordinate of ∈ H is added among the LIS, the root node that is about to all subtrees joins among the LIS, LSP then is changed to empty table,
Secondly, the element among the LIP is carried out importance scanning (Node Test), the element and the T that are about among the LIP compare, if element value then is important greater than T, output 1 moves into it among LSP then; Otherwise still be placed among the LIP, export 0 then.
Moreover importance scanning (Desc Test) is carried out in set to LIS.Maximum exports 0 less than T in a certain set; Otherwise export 1, then its inlet coordinate is deleted from LIS and cut apart rule according to set it is cut apart.If cut apart element that the new subclass of gained comprises more than one, the afterbody that then this subclass is added to LIS forms corresponding new list item, and simultaneously this element is Node Test; Only comprise an element if cut apart the new subclass of gained, then the Different Results that this element is tested according to its conspicuousness is added among the LIP respectively or the afterbody of LSP, and exports 0 or 1 respectively.
At last, the part element among the LSP (moving into that part of of LSP before importance scanning this time) is carried out refinement, element that promptly will refinement and T relatively as if then exporting this bit 1 greater than T, otherwise export 0.
Notice that decode procedure is with the cataloged procedure unanimity, as long as the output place is changed into input.
A key property in the SPIHT coding method is that sequencing information is not to transmit significantly, and the execution route of any this algorithm is all decided by the result of conspicuousness test.Successful decoding not only depends on current input, and NT (node tests) before depending on and the result of DT (descendent tests).Therefore single NT or the mistake of DT tend to lead to errors and are diffused into the entire image zone, thereby picture quality is caused serious destruction, this is the weak point of this algorithm just also, a promptly very little error code (such as the mistake of individual bit) just might make the entire image quality become very poor.This propagates in the channel of noise jamming is arranged is infeasible substantially, has also limited the extensive use of this algorithm to a great extent.
Summary of the invention
The present invention is directed to the defective or the deficiency that exist in the prior art, a kind of method of the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades is provided, adopt this method can effectively control the propagation on a large scale of error code mistake, strengthen the fault freedom of hierarchical tree set partitioning algorithm (SPIHT).
Technical conceive of the present invention is, on the basis of original hierarchical tree set partitioning algorithm (SPIHT), find and synchronization mechanism by piecemeal packing, mistake, be controlled at the inside of bag with form the mistake that causing is appearred in error code of bag, under the prerequisite that has increased redundant code stream and running time slightly, can control wrong propagation on a large scale effectively, thereby strengthen the fault freedom of this algorithm.
Technical scheme of the present invention is as follows:
The method of the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades, comprise coding method and corresponding coding/decoding method, it is characterized in that described coding method may further comprise the steps: input picture is carried out wavelet transformation, the lowest frequency zone that obtains behind the wavelet transformation is cut apart, form several n * n matrix, then direction in space tree set corresponding in each matrix is encoded, and the coding result of each matrix is packed;
Described each coding result formed packet infrastructure of packing is comprised a beginning flag, an information header, code stream section or imformosome, and an end mark;
Described information header comprises two parameters, and the number of importance node in its parameter 1 stored information bag, this parameter 1 can be in the importance scanning processes, by attached most importance to by differentiation at node will after the form that adds up obtain; The length of imformosome in its parameter 2 stored information bags, this parameter 2 can the form by the corresponding code stream length that adds up after output code flow obtain in the importance scanning process.
Described packing is meant to each coding result, will be used for that mistake prevents or the information of error-detecting, and the information of separating adds in each code stream that the coding back generates in the mode of head, and adds beginning flag and end mark, forms a bag.
Described coding/decoding method may further comprise the steps: read the additional information of bag, comprise beginning flag and information header, then the data of code stream section or imformosome are carried out importance scanning; The code stream section or the imformosome length value of the reality that obtains after the scanning are compared with information header relevant parameters value,, otherwise begin the mistake that error code causes is handled if equally then be left intact; Iteration is carried out this process, up to the importance been scanned of all bags.
Described coding/decoding method comprises that also the significant coefficient table that the data of bag code stream section or imformosome are carried out obtaining after the importance scanning carries out refinement scanning, if last scan did not occur wrong and did not arrive the Error Location, then be left intact, otherwise begin to carry out Synchronous Processing: skip the place of making a mistake, record is carried out in the position that mistake takes place when makeing mistakes, the number of the element of skipping obtains by the information header of erroneous packets.
Carry out following judgement behind the beginning flag reading bag, decision condition has two, and one is whether previous bag mistake took place, and another is whether beginning flag correct, these two conditions be or relation; Value with the number of the parameter importance node that reads when reading information header is stored in 2 variablees the Synchronous Processing when one of them is used for refinement.
The described mistake that error code is caused is handled and is comprised: the significant coefficient table of the current bag of (a) resetting, (b) write down the position of making a mistake, the inessential coefficient table and the inessential set table of the current bag of (c) resetting, (d) position, next packet header, location.
If mistake appears in code stream section or imformosome, directly synchronous by information header, if code stream section or imformosome and information header all make a mistake, then undertaken synchronously by bag starting and ending sign.
The position that mistake takes place is with in the middle of the data of the corresponding code stream length that should skip are kept at a structure simultaneously.
Technique effect of the present invention is as follows:
The present invention is on the basis of original coding method, be controlled at the inside of bag with form the mistake that causing is appearred in error code of bag, under the prerequisite that has increased redundant code stream and running time slightly, effectively controlled wrong propagation on a large scale, strengthened the fault freedom of this algorithm, and picture quality after rebuilding under the situation that error code do not occur and comparing after existing algorithm is rebuild do not change substantially.
The present invention compared with prior art has following characteristics:
The first, the lowest frequency subband is divided into the adjacency matrix of a series of n * n, direction in space tree set corresponding in each matrix is encoded.It is inner that the error code destruction that occurs causing just can be controlled at this set like this, and can not spread out.
The second, with result's packing of each direction in space tree collective encoding, error-detecting is included in the collective encoding information with the form of the key message that separates with head.Can find mistake and continue decoding by header like this, thereby reach the purpose that prevents error diffusion at next sync bit.
The 3rd, (as shown in Figure 3) simple in structure of bag, can effectively find mistake, and comprise 2 cover synchronization scenarios, and the code stream length that increases is very little.
Description of drawings
Fig. 1 is original SPIHT decoding method block diagram.
Fig. 2 is original SPIHT core encoder block diagram.
Fig. 3 is packet infrastructure figure of the present invention.
Fig. 4 decoding method block diagram of the present invention.
Fig. 5 is the realization flow figure of synchronization mechanism of the present invention in importance scanning.
Fig. 6 is the realization flow figure of synchronization mechanism of the present invention in refinement scanning.
Embodiment
The present invention is described in further detail below in conjunction with accompanying drawing.
A kind of improved SPIHT coding method comprises the wavelet transformation to input picture, calculates maximum and maximum matrix, initialization, importance scanning, five parts of refinement scanning.Wavelet transformation obtains wavelet coefficient matrix P; According to this matrix computations maximum and maximum matrix Q; Initialization iterations n is with the inessential coefficient table of form initialization (LIP), significant coefficient table (LSP), the inessential set table (LIS) of piecemeal; Importance scanning is carried out in direction in space tree set to the piece correspondence; The LSP table that whole importance scanning is obtained carries out refinement scanning.It is characterized in that at initial phase, (m * m) carry out piecemeal according to n * n size forms m/n group LIP, LSP, LIS table with the lowest frequency wavelet coefficient; When each group is carried out importance scanning, first output packet initial sum header packet information, and finishing this group importance scanning back output packet ending message.
For original hierarchical tree set partitioning algorithm (SPIHT), pack by piecemeal, wrong control is found and realized synchronously to mistake, the adjacency matrix that promptly when coding, the lowest frequency subband is divided into a series of n * n, direction in space tree set corresponding in each matrix is encoded, simultaneously some are used for mistake and prevent to add the code stream that generates behind each matrix coder in the mode of head, and add beginning flag and end mark to form a bag with the key message that separates., then synchronous again during decoding to encoding and decoding enforcement if find wrong bag, thus make the wrong bag that does not influence other, prevented the error code diffusion.
Adopt decoding method of the present invention to realize the examples of implementation of rest image codec such as Fig. 3～shown in Figure 6.
Packet infrastructure of the present invention as shown in Figure 3, it comprises a beginning flag FF4F, an information header A, a code stream section and an end mark FFD9.Information header A comprises 2 parameters, and parameter 1 is Num_IP, and the number of record important element can be in the importance scanning process, by attached most importance to by differentiation at node will after the form that adds up obtain; Parameter 2 is Num_Code, writes down the size of this bag, and unit is a bit, and this equally also can the form by the corresponding code stream length that adds up after output code flow obtain in the importance scanning process.The structure of this packets of information can play 2 layers synchronously, ground floor: if mistake appears in imformosome, and can be directly synchronous by information header; The second layer: if imformosome and information header all make a mistake, then can be undertaken synchronously by bag starting and ending sign, detailed process is seen Fig. 5 synchronously.
The realization block diagram of new rest image codec of the present invention as shown in Figure 4, its course of work is:
Coded portion
(1) entire image is carried out wavelet transformation, obtain a coefficient matrix P, calculate maximum by P then, generate a maximum matrix Q simultaneously, dimension is 1/2 of P, and (i j) is corresponding D (i, j) maximum in to element wherein.
(2) criterion (as 4 * 4) is cut apart according to certain in the lowest frequency zone that obtains behind the wavelet transformation and be divided into several (N) little square zone, represent with O (i).Each O (i) corresponding the set of one group of direction in space tree, according to the initialization rule in original spiht algorithm, use O (i) to i group LIP, LSP, the LIS table carries out initialization, just can obtain one and comprise N group LIP, LSP, the set LISP (i) that LIS shows.
(3) enter into block encoding after, add and unwrap beginning sign (FF4F), then add header packet information A, according to the scan method (as shown in Figure 2) identical O (i) is carried out importance scanning (only need in scanning process the parameter value of recording head information A get final product) then with original SPIHT, add end-of-packet sign (FFD9) behind the end of scan, so obtain wrapping C (i).Behind the whole set LISP importance end of scan, the element of making a difference that importance scanning before this is obtained carries out refinement scanning, and scan method is with the scan method of original SPIHT.Threshold value is reduced by half, and repeating step 3 is 0 up to the requirement of satisfying compression ratio or threshold value.
Decoded portion
(1) according to the same principle of coding to LIP, LSP, the LIS table carries out initialization, and obtains one and comprise N group LIP, LSP, the set LISP of LIS table.
(2) enter into piece decoding after, bag C (i) is carried out importance scanning, make a mistake if find bag C (i), then forward the importance scanning process to bag C (i+1) by synchronization mechanism (as shown in Figure 5), restart importance and scan.If do not make a mistake, then scanning is proceeded, and finishes up to C (N) decoding.Begin to carry out refinement scanning then, if mistake took place in last scanning, then will skip the place (as shown in Figure 6) of making a mistake when this refinement, record can be carried out in the position that mistake takes place when makeing mistakes, and the number of the element of skipping then can obtain by the header A of erroneous packets.If do not make a mistake, then scanning is proceeded.Threshold value is reduced by half, and repeating step 2 is 0 up to the requirement of satisfying compression ratio or threshold value.
The realization flow of synchronization mechanism of the present invention in importance scanning is described in detail as follows as shown in Figure 5:
(1) reads the beginning flag that wraps C (i),, then begin read head information if a last bag is not made mistakes or beginning flag equals FF4F.Otherwise begin to seek the position of satisfy condition (end-of-packet and bag beginning flag are ined succession) from a last packet header, promptly read 4 bytes successively continuously, if the result is FFD9FF4F, then correct packet header has been found in expression, otherwise continue down to seek, till satisfying condition.
(2) 2 of the read head information A parameters, and the important element of preserving current bag counts to NK, when being used for refinement synchronously.Begin importance scanning then.The parameter value of the reality of acquisition compared with the parameter value of the header that reads when importance was scanned, if identical, then proof does not make a mistake, and perhaps mistake does not influence the synchronous of encoding and decoding, continued importance scanning is carried out in next packet header.Otherwise enter the fault processing stage (3).
(3), at first the new element value that adds in the LSP table is changed to 0 again when entering fault processing after the stage; Write down the position PE that makes a mistake then; Then reinitialize LIP and LIS table, initialization procedure is the same with the preceding initialization of coding; At last, the position in the next packet header of packet header position calculation of parameter 2 (Num_Code) by header A and current bag, and flow process navigated to the Bao Touchu of next bag, continue new importance scanning process.
The realization flow of synchronization mechanism of the present invention in refinement scanning be as shown in Figure 6:
In refinement scanning, if the code stream that receives arrives Error Location PE, then by preserving corresponding information NK (each PE and corresponding NK are kept in the same structure), can skip the part of makeing mistakes, and the refinement flow process forwarded to correct position and begins new refinement scanning process, guarantee encoding and decoding synchronously.
Should be pointed out that the above embodiment can make those skilled in the art more fully understand the present invention, but do not limit the present invention in any way.Therefore, although this specification has been described in detail the present invention with reference to drawings and embodiments,, it will be appreciated by those skilled in the art that still and can make amendment or be equal to replacement the present invention; And all do not break away from the technical scheme and the improvement thereof of spirit of the present invention and technical spirit, and it all should be encompassed in the middle of the protection range of patent of the present invention.
Claims (7)
1. the method for the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades, comprise coding method and corresponding coding/decoding method, it is characterized in that described coding method may further comprise the steps: input picture is carried out wavelet transformation, the lowest frequency zone that obtains behind the wavelet transformation is cut apart, form several n * n matrix, then direction in space tree set corresponding in each matrix is encoded, and the coding result of each matrix is packed; Described each coding result formed packet infrastructure of packing is comprised a beginning flag, an information header, code stream section or imformosome, and an end mark; Described information header comprises two parameters, and the number of importance node in its parameter 1 stored information bag, this parameter 1 can be in the importance scanning processes, obtains with the form that adds up after wanting by being attached most importance to by differentiation at node; The length of code stream section or imformosome in its parameter 2 stored information bags, this parameter 2 can the form by the corresponding code stream length that adds up after output code flow obtain in the importance scanning process.
2. the method for the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades according to claim 1, it is characterized in that: described coding/decoding method may further comprise the steps: the additional information that reads bag, comprise beginning flag and information header, then the data of code stream section or imformosome are carried out importance scanning; The code stream section or the imformosome length value of the reality that obtains after the scanning are compared with information header relevant parameters value,, otherwise begin the mistake that error code causes is handled if equally then be left intact; Iteration is carried out this process, up to the importance been scanned of all bags.
3. the method for the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades according to claim 2, it is characterized in that: described coding/decoding method comprises that also the significant coefficient table that the data of bag code stream section or imformosome are carried out obtaining after the importance scanning carries out refinement scanning, if last scan did not occur wrong and did not arrive the Error Location, then be left intact, otherwise begin to carry out Synchronous Processing: skip the place of making a mistake, record is carried out in the position that mistake takes place when makeing mistakes, the number of the element of skipping obtains by the information header of erroneous packets.
4. the method for the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades according to claim 2, it is characterized in that: after reading the bag beginning flag, carry out following judgement, decision condition has two, one is whether previous bag mistake took place, another is whether beginning flag correct, these two conditions be or relation; Value with the number of the importance node that reads when reading information header is stored in 2 variablees the Synchronous Processing when one of them is used for refinement.
5. the method for the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades according to claim 2, it is characterized in that: the described mistake that error code is caused is handled and is comprised: the significant coefficient table of the current bag of (a) resetting, (b) write down the position of making a mistake, (c) the inessential coefficient table and the inessential set table of the current bag of replacement, (d) position, next packet header, location.
6. the method for the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades according to claim 2, it is characterized in that: if mistake appears in code stream section or imformosome, directly synchronous by information header, if code stream section or imformosome, all make a mistake with information header, then undertaken synchronously by bag starting and ending sign.
7. the method for the anti-error code capacity of tree collecting splitting algorithm coding-decoding device that upgrades according to claim 5 is characterized in that: the position that mistake takes place is with in the middle of the data of the corresponding code stream length that should skip are kept at a structure simultaneously.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100118672A CN100435550C (en) | 2006-05-10 | 2006-05-10 | Method for increasing anti-bit-error ability of grade-tree collecting splitting algorithm coding-decoding device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100118672A CN100435550C (en) | 2006-05-10 | 2006-05-10 | Method for increasing anti-bit-error ability of grade-tree collecting splitting algorithm coding-decoding device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1852391A CN1852391A (en) | 2006-10-25 |
| CN100435550C true CN100435550C (en) | 2008-11-19 |
Family
ID=37133857
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100118672A Expired - Fee Related CN100435550C (en) | 2006-05-10 | 2006-05-10 | Method for increasing anti-bit-error ability of grade-tree collecting splitting algorithm coding-decoding device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100435550C (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102984519A (en) * | 2012-11-22 | 2013-03-20 | 西安电子科技大学 | Image coding method for static and dynamic integration |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030026488A1 (en) * | 2001-05-30 | 2003-02-06 | Keun-Hyeong Park | Method of coding image data using region of interest |
| CN1581977A (en) * | 2004-05-20 | 2005-02-16 | 复旦大学 | Tree-structure-based grade tree aggregation-divided video image compression method |
-
2006
- 2006-05-10 CN CNB2006100118672A patent/CN100435550C/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030026488A1 (en) * | 2001-05-30 | 2003-02-06 | Keun-Hyeong Park | Method of coding image data using region of interest |
| CN1581977A (en) * | 2004-05-20 | 2005-02-16 | 复旦大学 | Tree-structure-based grade tree aggregation-divided video image compression method |
Non-Patent Citations (6)
| Title |
|---|
| Embedded and Scalable Image Coding Based onVirtualSPIHT. Chen Hongxin, Liu Zhengguang.The 8th International Conference on Advanced Communication Technology,Vol.1 . 2006 |
| Embedded and Scalable Image Coding Based onVirtualSPIHT. Chen Hongxin, Liu Zhengguang.The 8th International Conference on Advanced Communication Technology,Vol.1 . 2006 * |
| SPIHT算法容错性能的分析及改进. 廖小涛，张晓林，刘荣科.遥测遥控，第24卷第5期. 2003 |
| SPIHT算法容错性能的分析及改进. 廖小涛，张晓林，刘荣科.遥测遥控，第24卷第5期. 2003 * |
| 一种抗误码的SPIHT改进算法. 祝新平，樊桂花，李迎春.第十二届全国图象图形学学术会议论文集. 2005 |
| 一种抗误码的SPIHT改进算法. 祝新平，樊桂花，李迎春.第十二届全国图象图形学学术会议论文集. 2005 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1852391A (en) | 2006-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wayner | Mimic functions | |
| CN102143367B (en) | Method, device and system for checking error correction | |
| Alatan et al. | Unequal error protection of SPIHT encoded image bit streams | |
| CN102131083B (en) | Method and device for transmitting scalable videos based on priorities | |
| CN102687445B (en) | Forward error correction encoding,decoding method,apparatus and system | |
| CN107276722A (en) | A kind of data transmission method and system based on UDP | |
| US20080052605A1 (en) | Method for Restoring the Missing Data Packets | |
| CN100479524C (en) | Package for FGS/PFGS video frequency potential flow | |
| CN103858433A (en) | Hierarchical entropy encoding and decoding | |
| CN101087181B (en) | A method for removing interweaving and speed match | |
| CN101432969A (en) | Forward error-correcting (FEC) coding and streaming | |
| WO2008060025A1 (en) | Method and system for partitioning and encoding of uncompressed video for transmission over wireless medium | |
| US7464319B2 (en) | Forward error correction with codeword cross-interleaving and key-based packet compression | |
| CN104081701B (en) | Communication means, system and the device of optical network system | |
| CN108966259B (en) | Anti-interference transmission method based on network coding | |
| CN111917512A (en) | High-efficiency coding design method based on BATS code in multi-hop transmission system | |
| CN106656205A (en) | Polarization code decoding method and system capable of reducing memory consumption | |
| Rudow et al. | Streaming codes for variable-size arrivals | |
| Wunderlich et al. | We don't need no generation-a practical approach to sliding window RLNC | |
| Baruffa et al. | Error protection and interleaving for wireless transmission of JPEG 2000 images and video | |
| CN1656692A (en) | Apparatus for iterative hard-decision forward error correction decoding | |
| CN107896137A (en) | A kind of sort method suitable for the division of polarization code decoding path | |
| CN100435550C (en) | Method for increasing anti-bit-error ability of grade-tree collecting splitting algorithm coding-decoding device | |
| US8514895B2 (en) | Bit-stuffing method for crosstalk avoidance in high-speed buses | |
| CN102325252B (en) | Optimal-wavelet-packet-based joint source channel coding method and image transmission system |
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 | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081119 Termination date: 20110510 |