CN104349165A - High-performance length-variable coding and decoding methods and devices - Google Patents
High-performance length-variable coding and decoding methods and devices Download PDFInfo
- Publication number
- CN104349165A CN104349165A CN201410469348.5A CN201410469348A CN104349165A CN 104349165 A CN104349165 A CN 104349165A CN 201410469348 A CN201410469348 A CN 201410469348A CN 104349165 A CN104349165 A CN 104349165A
- Authority
- CN
- China
- Prior art keywords
- golomb
- value
- decoding
- pixel
- residual error
- 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
Landscapes
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention discloses high-performance length-variable coding and decoding methods and devices. A mapping value of a pixel point residual error is coded by using a coding device at a coding end. A specific coding method comprises the following steps: firstly, coding a Golomb coefficient Q of the mapping value of the pixel point residual error by using a Huffman coding method; secondly, coding a Golomb remainder R of the mapping value of the pixel point residual error by using a Golomb coding method; decoding a bit stream by using a decoding method at a decoding end. A specific decoding method comprises the following steps: firstly, reading the bit streams of a plurality of pixel points within a clock period; secondly, decoding Q and R of each pixel point in parallel; finally, calculating the mapping value of the residual error of each pixel point in parallel according to the Q and the R. According to the high-performance length-variable coding and decoding methods and devices, the problems of low coding efficiency in an existing length-variable coding method, and long time delay and large area of a decoding device and the like can be effectively solved.
Description
Technical field
The present invention, about a kind of decoding method and device, particularly relates to high-performance variable first draft coding/decoding method and the device of a kind of video or image processing field.
Background technology
Image or video are a kind of effectively lively information representation modes, obtain at present use widely in fields such as TV, video conference and DVD.In order to solve the shortcomings such as image or cost of video storage height and transport tape are roomy, linked groups has formulated video compression standard as MPEG-x series and image compression standard JPEG series etc. in the world.Main code method in image or video compression standard has that Huffman encodes, Golomb-Rice coding and arithmetic coding, makes code efficiency higher.
Huffman encodes: Huffman coding is a kind of harmless entropy code mode, and its main coding thinking is: to the higher information distribution of probability of occurrence with less code word, and the less information distribution of probability of occurrence is with more code word.Its shortcoming is: when information category many (as 256 kinds) and probability distribution is even time, compression efficiency will be very low, and decoding device by multistage MUX, not only need produce huge delay and make decoding device area increase simultaneously.
Golomb-Rice encodes: Golomb-Rice coding is a kind of harmless entropy code mode, and its main code mode is: according to the K value preset, first calculate coefficient Q, coding Q individual 0; Then encode a bit flag position M, and now M value must be 1; Finally use K position coded remainder R.Its shortcoming is: when the K value preset is inaccurate, compression efficiency will be very low, and decoding device by multistage MUX, not only need produce huge delay and make decoding device area increase simultaneously.
Summary of the invention
For overcoming the deficiency that above-mentioned prior art exists, the object of the present invention is to provide a kind of high-performance variable first draft coding/decoding method and device, the problem that it extends when efficiently solving the low and decoding device of code efficiency in existing variable length encoding method, area is large etc., achieves high-performance variable-length code decode.
For reaching above-mentioned and other object, the present invention proposes a kind of high-performance variable length encoding method, comprises the steps:
Step one, according to current line pixel value and encoded pixel value, calculates predicted value;
Step 2, calculates residual values according to this predicted value and current pixel value and is mapped as nonnegative integer;
Step 3, calculates the K value needed for Golomb-Rice coding according to the information of encoded pixel value;
Step 4, adopts the Golomb coefficient Q of the mapping value of Huffman encoding method encodes pixel residual error, then adopts the Golomb remainder R of the mapping value of Golomb encoding method encodes pixel residual error, encodes to all residual error mapping value to be encoded;
Step 5, carries out packing to the avriable length codes produced after coding and produces final bit stream.
Further, step 4 comprises the steps: further
Step 2.1, utilizes residual error mapping value computing formula to calculate Golomb coefficient Q and the Golomb remainder R of residual error mapping value to be encoded according to K value;
Step 2.2, the Golomb coefficient Q of the residual error mapping value adopting Huffman coded system to encode to be encoded;
Step 2.3, adopts Golomb coding method and the Golomb remainder R of the residual error mapping value utilizing binary coding to be encoded;
Step 2.4, produces this Golomb coefficient Q and the polymerization of this Golomb remainder R the avriable length codes of residual error mapping value to be encoded;
Step 2.5, the residual error mapping value to be encoded that circulation step 2.1 ~ 34 coding is all.
Further, the residual error mapping value computing formula in step 2.1 is:
dm＝Q×2
k+R
Wherein, dm is residual error mapping value to be encoded, and Q is Golomb coefficient Q, R is Golomb remainder R.
Further, in step 2.2, this Huffman coded system is to encode this Golomb coefficient Q value according to the code table that pre-establishes.
For achieving the above object, the present invention also provides a kind of high-performance variable length decoding method, comprises the steps:
Step one, carries out the longest bit stream needed for de-packaging operation generation Golomb-TT decoding device to former bit stream;
Step 2, goes out predicted value according to current line pixel value and encoded calculated for pixel values;
Step 3, calculates the K value needed for Golomb-Rice decoding according to the information of encoded pixel value;
Step 4, utilize Golomb coefficient Q and the Golomb remainder R of the multiple pixel of this Golomb-TT decoding device parallel decoding, and according to K value and Golomb coefficient Q and Golomb remainder R, utilize residual error mapping value computing formula to calculate the residual error mapping value of each pixel;
Step 5, carries out reflection according to predicted value and current residue mapping value and penetrates and calculate current pixel value.
Further, in step 4, adopt Huffman decoding process decoding Golomb coefficient Q, from bit stream, decode binary Golomb remainder R value, and according to K, Q and R value, utilize this residual error mapping value computing formula to calculate the residual error mapping value of each pixel.
Further, this residual error mapping value computing formula is:
dm＝Q×2
k+R
Wherein, dm is residual error mapping value to be decoded, and Q is Golomb coefficient Q, R is Golomb remainder R.
For achieving the above object, the present invention also provides a kind of high-performance variable long decode device, at least comprises:
Parse module, carries out the longest bit stream needed for de-packaging operation generation Golomb-TT decoding device to former bit stream;
Prediction module, goes out predicted value according to current line pixel value and encoded calculated for pixel values;
K value computing module, calculates the K value needed for Golomb-Rice decoding according to the information of this encoded pixel value;
Golomb-TT decoding device, the bit stream of three pixels is at least read within a clock cycle, the parallel Huffman decoding process that adopts is decoded the Golomb coefficient Q of each pixel, from bit stream, decode the Golomb remainder R of binary each pixel, and go out the mapping value of the residual error of each pixel according to K value, Golomb coefficient Q and the parallel computation of Golomb remainder R;
Anti-mapping block, carries out reflection according to predicted value pix_pred and current residue mapping value and penetrates, calculate current pixel value.
Further, this Golomb-TT decoding device adopts the Huffman code table in dichotomy decoding Huffman decoding process, to reduce the progression of the MUX in decoding Golomb coefficient Q.
Further, this Golomb-TT decoding device adopts parallel mode to decode within a clock cycle multiple pixel component, and the bit number that parallel addition (n-1)th pixel component consumes when the Golomb coefficient Q of decoding n-th pixel component.
Compared with prior art, the present invention's a kind of high-performance variable first draft coding/decoding method and device are by first adopting the Golomb coefficient (Quotient of the mapping value of Huffman encoding method encodes pixel residual error, Q), then the Golomb remainder (Remainder of the mapping value of Golomb encoding method encodes pixel residual error is adopted, R), and adopt dichotomy to reduce the progression of the required MUX of Huffman decoding when decoding end is decoded, the problem extended when simultaneously adopting parallel mode to solve, extend when efficiently solving the low and decoding device of code efficiency in existing variable length encoding method, the problem of the large grade of area.
Accompanying drawing explanation
Fig. 1 is the flow chart of steps of a kind of high-performance variable length encoding method of the present invention;
Fig. 2 is an application scenarios schematic diagram of a kind of high-performance variable length encoding method of the present invention;
Fig. 3 is the thin portion flow chart of step 104 in a kind of high-performance of the present invention variable length encoding method;
Fig. 4 is the flow chart of steps of a kind of high-performance variable length decoding method of the present invention;
Fig. 5 is the system architecture diagram of a kind of high-performance variable long decode of the present invention device;
Fig. 6 is that in the present invention, each clock is decoded the waveform schematic diagram of dm value of 3 components;
Fig. 7 is longest path schematic diagram when adopting the Q value in conventional method decoding Golomb-TT decoding device;
Fig. 8 is longest path schematic diagram when decoding Q value in Golomb-TT decoding device of the present invention
Fig. 9 is the parallel decoding processes schematic diagram of Golomb-TT decoding device proposed by the invention
Figure 10 is that the present invention adopts parallel mode to solve the schematic diagram of delay problem.
Below by way of specific instantiation and accompanying drawings embodiments of the present invention, those skilled in the art can understand other advantage of the present invention and effect easily by content disclosed in the present specification.The present invention is also implemented by other different instantiation or is applied, and the every details in this specification also can based on different viewpoints and application, carries out various modification and change not deviating under spirit of the present invention.
Before explaining the present invention, first defined declaration is carried out to several nouns involved in the present invention:
(1) map: establish A, B to be two nonempty sets, if there is a rule F, make each element a in A, by rule F, have well-determined element b corresponding with it in B, then title F is the mapping from A to B;
(2) residual values: the difference of predicted value and currency;
(3) MUX: in multichannel data transport process, the circuit that can as required wherein any road be elected;
(4) dichotomy: the method be divided into two;
Fig. 1 is the flow chart of steps of a kind of high-performance variable length encoding method of the present invention, and Fig. 2 is an application scenarios schematic diagram of a kind of high-performance variable length encoding method of the present invention.As shown in Figures 1 and 2, a kind of high-performance variable length encoding method of the present invention, comprises the steps:
Step 101, according to the encoded pixel values such as current line pixel value and lastrow (f (x-1, x-2 ... y-1, y-2 ... )) calculate predicted value (pix_pred).
Step 102, calculates residual values according to pix_pred and current pixel value (pix_cur) and is mapped as nonnegative integer (diffmap, dm), such as residual values be-255 ~ 255 dm value be 0 ~ 510.
Step 103, according to the encoded pixel value f such as lastrow (x-1, x-2 ... y-1, y-2 ...) information calculate Golomb-Rice coding needed for K value.
Step 104, adopt the Golomb coefficient (Quotient of the mapping value dm of Huffman encoding method encodes pixel residual error, Q), then the Golomb remainder (Remainder of the mapping value dm of Golomb encoding method encodes pixel residual error is adopted, R), all dm values to be encoded are encoded; For sake of convenience, step 104 is referred to as Golomb-TT coding method.
Step 105, carries out packing to the avriable length codes (code word, cw) produced after coding and produces final bit stream (bitstream, bs).
Table 1 is the contrast table of code word and the required bit number produced after existing block code method and traditional Golomb-Rice (K==2) coding method and Golomb-TT proposed by the invention (K==2) coding method are encoded to dm value, wherein " x " represents that value can be 0 or 1, { y} represents this code word size, and y can all nonnegative integers of value.In table 1, concrete meaning is:
First row: four numbers are the index of the dm value of a group;
Secondary series: residual values for-255 ~ 255 time dm value institute's likely value, wherein front 127 behavior 4 number one group, last column i.e. 128 behavior 2 number one group;
3rd row: the code word of block code and code word size;
4th row: the code word of Golomb-Rice coding and code word size;
5th row: the code word of Golomb-TT coding and code word size;
The Golomb-TT coding method that table 1 is proposed by the invention and existing coding method contrast
| Index value | Dm value | Block code | Golomb-Rice(K＝2) | Golomb-TT(K＝2) |
| 1 | 0～3 | 00000000xx{9} | 1xx{3} | 1xx{3} |
| 2 | 4～7 | 00000001xx{9} | 01xx{4} | 010xx{5} |
| 3 | 8～11 | 00000010xx{9} | 001xx{5} | 011xx{5} |
| 4 | 12～15 | 00000011xx{9} | 0001xx{6} | 0010xx{6} |
| 5 | 16～19 | 00000100xx{9} | 00001xx{7} | 0011xx{6} |
| 6 | 20～23 | 00000101xx{9} | 000001xx{8} | 00010xx{7} |
| 7 | 24～27 | 00000110xx{9} | 0000001xx{9} | 00011xx{7} |
| 8 | 28～31 | 00000111xx{9} | 00000001xx{10} | 000000xx{8} |
| 9 | 32～35 | 00001000xx{9} | 000000001xx{11} | 000001xx{8} |
| 10 | 36～39 | 00001001xx{9} | 0000000001dm value { 19} | 000010xx{8} |
| 11 | 40～43 | 00001010xx{9} | 0000000001dm value { 19} | 000011dm value { 15} |
| … | … | … | … | … |
| 127 | 504～507 | 1111110xx{9} | 0000000001dm value { 19} | 000011dm value { 15} |
| 128 | 508～510 | 1111111xx{9} | 0000000001dm value { 19} | 000011dm value { 15} |
As can be seen from Table 1, when index value is 2, the required code word size of Golomb-Rice coding is the required code word size of 4, Golomb-TT coding is 5, and in all the other situations, the required code word size of Golomb-TT coding is all shorter than Golomb-Rice.
Table 2 is that the code word that produces after traditional Golomb-Rice (K==3) coding method and Golomb-TT proposed by the invention (K==3) coding method are encoded to dm value and required bit number contrast.
Golomb-TT (K=3) coding method that table 2 is proposed by the invention and Golomb-Rice (K=3) coding method contrast
| Index value | Dm value | Golomb-Rice(K＝3) | Golomb-TT(K＝3) |
| 1 | 0～7 | 1xxx{4} | 1xxx{4} |
| 2 | 8～15 | 01xxx{5} | 010xxx{6} |
[0074]
| 3 | 16～23 | 001xxx{6} | 011xxx{6} |
| 4 | 24～31 | 0001xxx{7} | 0010xxx{7} |
| 5 | 32～39 | 00001xxx{8} | 0011xxx{7} |
| 6 | 40～47 | 000001xxx{9} | 00010xxx{8} |
| 7 | 48～55 | 0000001xxx{10} | 00011xxx{8} |
| 8 | 56～63 | 00000001xxx{11} | 000000xxx{9} |
| 9 | 64～71 | 000000001xxx{12} | 000001xxx{9} |
| 10 | 72～79 | 0000000001dm value { 19} | 000010xxx{9} |
| 11 | 80～87 | 0000000001dm value { 19} | 000011dm value { 15} |
| … | … | … | … |
| 63 | 496～503 | 0000000001dm value { 19} | 000011dm value { 15} |
| 64 | 504～510 | 0000000001dm value { 19} | 000011dm value { 15} |
As can be seen from Table 2, when index value is 2, the required code word size of Golomb-Rice coding is the required code word size of 5, Golomb-TT coding is 6, and in all the other situations, the required code word size of Golomb-TT coding is all shorter than Golomb-Rice.
Draw to draw a conclusion from table 1,2 results reasoning:
Conclusion 1: when K value is less than or equal to 3, Golomb-TT coding method proposed by the invention has the probability of 98% to be better than Golomb-Rice coding method;
Conclusion 2: when K value prediction is accurate, Golomb-TT code efficiency proposed by the invention and Golomb-Rice maintain an equal level;
Conclusion 3: when K value prediction is inaccurate, Golomb-TT code efficiency proposed by the invention will be better than Golomb-Rice greatly;
Fig. 3 is the thin portion flow chart of step 104 in a kind of high-performance of the present invention variable length encoding method.As shown in Figure 3, the Golomb-TT coding method of step 104 comprises the steps: further
Step 301, calculate Q and the R value of dm value to be encoded according to K value, specific formula for calculation is as " eq1 ";
dm＝Q×2
k+R (eq1)
Step 302, adopts Huffman coded system coding Q value, namely according to the code table pre-established (code table used in table 1,2 is see table 3) coding Q value, illustrates: this code table can adjust according to the concrete distribution probability of information.
Table 3 is encoded a kind of Huffman code table of Q value
| Q | Code word |
| 0 | 1 |
| 1 | 010 |
| 2 | 011 |
[0086]
| 3 | 0010 |
| 4 | 0011 |
| 5 | 00010 |
| 6 | 00011 |
| 7 | 000000 |
| 8 | 000001 |
| 9 | 000010 |
| other | 000011 |
Step 303, adopts Golomb coding method and the R value of the dm value utilizing binary coding to be encoded.
Step 304, produces Q and R polymerization the avriable length codes cw of dm value to be encoded.
Step 305, the dm value to be encoded that circulation step 301 ~ 304 coding is all.
For the further comparative illustration of the dm value Golomb-TT coding method proposed by the invention that index value in table 2 is 3.As known from Table 2, the span of dm is 16 ~ 23, can calculate its Q value and be " 2 ", and the span of R is 0 ~ 7 according to formula " eq1 "; Then show that the corresponding code word of Q is for " 011 " according to the code table shown in table 3; Then be " xxx " by binary representation R value, wherein the value of " x " is " 0 " or " 1 "; Finally the polymerization of the code word " xxx " of the code word of Q " 011 " and R is produced code word cw " 011xxx ", illustrates: in cw Q and R can carry out multi-form sequence according to real needs, another common sortord is " xxx011 ".Fig. 4 be adopt in this example index value in Golomb-TT coded system of the present invention and Golomb-Rice coded system his-and-hers watches 2 be 3 dm value to carry out encoding the comparison diagram of schematic diagram.
Fig. 5 is the flow chart of steps of a kind of high-performance variable length decoding method of the present invention.As shown in Figure 4, a kind of high-performance variable length decoding method of the present invention, comprises the steps:
Step 501, carrying out the longest bit stream needed for de-packaging operation generation Golomb-TT decoding device possibility to former bit stream (bs), such as, is bs16 in the present embodiment.
Step 502, according to the encoded pixel values such as current line pixel value and lastrow (f (x-1, x-2 ... y-1, y-2 ... )) calculate predicted value (pix_pred).
Step 503, according to the encoded pixel value f such as lastrow (x-1, x-2 ... y-1, y-2 ...) information calculate Golomb-Rice decoding needed for K value.
Step 504, adopts Huffman decoding process decoding Q value, namely according to the code table decoding Q value pre-established, decodes binary R value, and according to K, Q and R value, utilize aforementioned formula " eq1 " to calculate dm value from bit stream (bs).
Step 505, carries out reflection according to predicted value pix_pred and current dm value and penetrates and calculate current pixel value pix_cur.
Fig. 6 is the system architecture diagram of a kind of high-performance variable long decode of the present invention device, as shown in Figure 6, a kind of high-performance variable long decode of the present invention device, comprising: parse module 601, prediction module 602, K value computing module 603, Golomb-TT decoding device 604 and anti-mapping block 605.
Wherein, parse module 601 carries out the longest bit stream needed for de-packaging operation generation Golomb-TT decoding device possibility to former bit stream (bs), such as, is bs16 in the present embodiment; Prediction module 602, according to the encoded pixel values such as current line pixel value and lastrow (f (x-1, x-2 ... y-1, y-2 ... )) calculate predicted value (pix_pred); K value computing module 603, according to the encoded pixel value f such as lastrow (x-1, x-2 ... y-1, y-2 ...) information calculate Golomb-Rice decoding needed for K value; Golomb-TT decoding device 604 adopts Huffman decoding process decoding Q value, namely according to the code table decoding Q value pre-established, from bit stream (bs), decode binary R value, and according to K, Q and R value, utilize aforementioned formula " eq1 " to calculate dm value; Anti-mapping block 605 carries out reflection according to predicted value pix_pred and current dm value and penetrates and calculate current pixel value pix_cur.
In hardware implementing, often need the factors such as balance area, efficiency and power consumption.In Video post-processing process, first compression is carried out to video source and the code stream after compression is sent into chip external memory in a lot of situation, when needs, from chip external memory, readout code stream decompresses, because Video post-processing technology has multiple as noise reduction, FRC and enhancing etc., so need the decompressing device of multichannel, and each road decompressing device must meet the operating efficiency required by Video post-processing technology, this just proposes the requirement of area and efficiency to the decoding device of one of main modular in decompressing device.Such as the video of YUV444 form, the present invention's two kinds of proposing improve each clock of method (clock) of decoding devices and can to decode 3 luminance components (compenont), namely each clock of Golomb-TT decoding device is needed to decode 3 dm values, as shown in Figure 7.
In Golomb-Rice and Golomb-TT decoding device, the path that time delay is the longest is the decoded portion to Q value, use conventional methods, the former need by 9 grades of MUX, the latter need by 6 grades of MUX, Fig. 8 is that the latter adopts conventional method to decode the latency path figure of Q value, for six bit stream bs [0]-bs [5], the longest from starting point sp to terminal ep need through 6 grades of MUX.
The present invention proposes the improvement to method as shown in Figure 8, namely similar dichotomy is adopted to decode to the Q in Golomb-TT decoding device, adopt the present invention propose improve one's methods after, Golomb-TT decoding device 504 is decoded the Q value in Golomb-TT, need through 4 grades of MUX from starting point sp to the longest of terminal ep, as shown in Figure 9.In present pre-ferred embodiments, the Huffman code table that Golomb-TT decoding device adopts dichotomy to decode proposed by the invention, its concrete steps are:
Step1: first judge low 4 whether entirely as 0;
Step2: then walk abreast and judge minimum sum-bit the 5th;
Step3: then walk abreast and judge first and second and five;
Step4: finally judge third and fourth and five.
In high performance variable length decoding device, require to decode multiple pixel component within a clock cycle, such as 3 components.With traditional method for designing design Golomb-TT decoding device time, time elongatedness be: 6 adders of the MUX of 18 grades and 32 ends.The present invention adopts parallel mode to solve delay problem, specifically as shown in Figure 10, Q and R (dec_Q0, dec_Q1, dec_Q2 of Golomb-TT decoding device three components of parallel decoding bit stream bs_48 within a clock cycle, dec_R0, dec_R1, dec_R2), the bit number len_y0 that parallel addition one-component Q0 consumes when decoding second component Q1, the bit number len_y1 that parallel addition second component consumes when decoding the 3rd component Q2.Adopt the Golomb-TT decoding device of the present invention to carry out Golomb-TT decoding, then time, elongatedness is: 6 adders of the MUX of 12 grades and 12 end.
In sum, the present invention's a kind of high-performance variable first draft coding/decoding method and device are by first adopting the Golomb coefficient (Quotient of the mapping value of Huffman encoding method encodes pixel residual error, Q), then the Golomb remainder (Remainder of the mapping value of Golomb encoding method encodes pixel residual error is adopted, R), and adopt dichotomy to reduce the progression of the required MUX of Huffman decoding when decoding end is decoded, the problem extended when simultaneously adopting parallel mode to solve, extend when efficiently solving the low and decoding device of code efficiency in existing variable length encoding method, the problem of the large grade of area.
Above-described embodiment is illustrative principle of the present invention and effect thereof only, but not for limiting the present invention.Any those skilled in the art all without prejudice under spirit of the present invention and category, can carry out modifying to above-described embodiment and change.Therefore, the scope of the present invention, should listed by claims.
Claims (10)
1. a high-performance variable length encoding method, comprises the steps:
Step one, according to current line pixel value and encoded pixel value, calculates predicted value;
Step 2, calculates residual values according to this predicted value and current pixel value and is mapped as nonnegative integer;
Step 3, calculates the K value needed for Golomb-Rice coding according to the information of encoded pixel value;
Step 4, adopts the Golomb coefficient Q of the mapping value of Huffman encoding method encodes pixel residual error, then adopts the Golomb remainder R of the mapping value of Golomb encoding method encodes pixel residual error, encodes to all residual error mapping value to be encoded;
Step 5, carries out packing to the avriable length codes produced after coding and produces final bit stream.
2. a kind of high-performance variable length encoding method as claimed in claim 1, it is characterized in that, step 4 comprises the steps: further
Step 2.1, utilizes residual error mapping value computing formula to calculate Golomb coefficient Q and the Golomb remainder R of residual error mapping value to be encoded according to K value;
Step 2.2, the Golomb coefficient Q of the residual error mapping value adopting Huffman coded system to encode to be encoded;
Step 2.3, adopts Golomb coding method and the Golomb remainder R of the residual error mapping value utilizing binary coding to be encoded;
Step 2.4, produces this Golomb coefficient Q and the polymerization of this Golomb remainder R the avriable length codes of residual error mapping value to be encoded;
Step 2.5, the residual error mapping value to be encoded that circulation step 2.1 ~ 34 coding is all.
3. a kind of high-performance variable length encoding method as claimed in claim 2, is characterized in that, the residual error mapping value computing formula in step 2.1 is:
dm＝Q×2
k+R
Wherein, dm is residual error mapping value to be encoded, and Q is Golomb coefficient Q, R is Golomb remainder R.
4. a kind of high-performance variable length encoding method as claimed in claim 2, is characterized in that: in step 2.2, and this Huffman coded system is to encode this Golomb coefficient Q value according to the code table that pre-establishes.
5. a high-performance variable length decoding method, comprises the steps:
Step one, carries out the longest bit stream needed for de-packaging operation generation Golomb-TT decoding device to former bit stream;
Step 2, goes out predicted value according to current line pixel value and encoded calculated for pixel values;
Step 3, calculates the K value needed for Golomb-Rice decoding according to the information of encoded pixel value;
Step 4, utilize Golomb coefficient Q and the Golomb remainder R of the multiple pixel of this Golomb-TT decoding device parallel decoding, and according to K value and Golomb coefficient Q and Golomb remainder R, utilize residual error mapping value computing formula to calculate the residual error mapping value of each pixel;
Step 5, carries out reflection according to predicted value and current residue mapping value and penetrates and calculate current pixel value.
6. a kind of high-performance variable length decoding method as claimed in claim 5, it is characterized in that: in step 4, adopt Huffman decoding process decoding Golomb coefficient Q, binary Golomb remainder R value is decoded from bit stream, and according to K, Q and R value, utilize this residual error mapping value computing formula to calculate the residual error mapping value of each pixel.
7. a kind of high-performance variable length decoding method as claimed in claim 6, is characterized in that: this residual error mapping value computing formula is:
dm＝Q×2
k+R
Wherein, dm is residual error mapping value to be decoded, and Q is Golomb coefficient Q, R is Golomb remainder R.
8. a high-performance variable long decode device, at least comprises:
Parse module, carries out the longest bit stream needed for de-packaging operation generation Golomb-TT decoding device to former bit stream;
Prediction module, goes out predicted value according to current line pixel value and encoded calculated for pixel values;
K value computing module, calculates the K value needed for Golomb-Rice decoding according to the information of this encoded pixel value;
Golomb-TT decoding device, the bit stream of three pixels is at least read within a clock cycle, the parallel Huffman decoding process that adopts is decoded the Golomb coefficient Q of each pixel, from bit stream, decode the Golomb remainder R of binary each pixel, and go out the mapping value of the residual error of each pixel according to K value, Golomb coefficient Q and the parallel computation of Golomb remainder R;
Anti-mapping block, carries out reflection according to predicted value pix_pred and current residue mapping value and penetrates, calculate current pixel value.
9. a kind of high-performance variable long decode device as claimed in claim 8, it is characterized in that: this Golomb-TT decoding device adopts the Huffman code table in dichotomy decoding Huffman decoding process, to reduce the progression of the MUX in decoding Golomb coefficient Q.
10. a kind of high-performance variable long decode device as claimed in claim 9, it is characterized in that: this Golomb-TT decoding device adopts parallel mode to decode within a clock cycle multiple pixel component, and the bit number that parallel addition (n-1)th pixel component consumes when the Golomb coefficient Q of decoding n-th pixel component.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410469348.5A CN104349165B (en) | 2014-09-16 | 2014-09-16 | The elongated decoding method of high-performance and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410469348.5A CN104349165B (en) | 2014-09-16 | 2014-09-16 | The elongated decoding method of high-performance and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104349165A true CN104349165A (en) | 2015-02-11 |
| CN104349165B CN104349165B (en) | 2017-06-16 |
Family
ID=52503848
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410469348.5A Active CN104349165B (en) | 2014-09-16 | 2014-09-16 | The elongated decoding method of high-performance and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104349165B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105915227A (en) * | 2016-04-08 | 2016-08-31 | 苏州大学 | Adaptive mixed data lossless compression system |
| CN105915228A (en) * | 2016-04-08 | 2016-08-31 | 苏州大学 | Adaptive mixed data lossless compression method |
| CN108155969A (en) * | 2016-12-02 | 2018-06-12 | 北京深鉴科技有限公司 | A kind of decoding circuit for variable length coded data |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070282933A1 (en) * | 2006-06-05 | 2007-12-06 | Donald Martin Monro | Data coding |
| US20100296746A1 (en) * | 2008-01-21 | 2010-11-25 | Telefonaktiebolaget Lm Ericsson (Publ) | Prediction-Based Image Processing |
| CN102215385A (en) * | 2010-04-09 | 2011-10-12 | 中国科学院沈阳自动化研究所 | Real-time lossless compression method for image |
| CN102368385A (en) * | 2011-09-07 | 2012-03-07 | 中科开元信息技术(北京)有限公司 | Backward block adaptive Golomb-Rice coding and decoding method and apparatus thereof |
| CN102422540A (en) * | 2009-05-20 | 2012-04-18 | 日本电信电话株式会社 | Coding method, coding device, decoding method, decoding device, program, and recording medium |
| CN102595496A (en) * | 2012-03-08 | 2012-07-18 | 西北大学 | Context-adaptive quotient and remainder encoding method used for sensing data of wireless sensing nodes |
| CN103702129A (en) * | 2012-09-27 | 2014-04-02 | 联咏科技股份有限公司 | Video coding method and video coding device |
-
2014
- 2014-09-16 CN CN201410469348.5A patent/CN104349165B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070282933A1 (en) * | 2006-06-05 | 2007-12-06 | Donald Martin Monro | Data coding |
| US20100296746A1 (en) * | 2008-01-21 | 2010-11-25 | Telefonaktiebolaget Lm Ericsson (Publ) | Prediction-Based Image Processing |
| CN102422540A (en) * | 2009-05-20 | 2012-04-18 | 日本电信电话株式会社 | Coding method, coding device, decoding method, decoding device, program, and recording medium |
| CN102215385A (en) * | 2010-04-09 | 2011-10-12 | 中国科学院沈阳自动化研究所 | Real-time lossless compression method for image |
| CN102368385A (en) * | 2011-09-07 | 2012-03-07 | 中科开元信息技术(北京)有限公司 | Backward block adaptive Golomb-Rice coding and decoding method and apparatus thereof |
| CN102595496A (en) * | 2012-03-08 | 2012-07-18 | 西北大学 | Context-adaptive quotient and remainder encoding method used for sensing data of wireless sensing nodes |
| CN103702129A (en) * | 2012-09-27 | 2014-04-02 | 联咏科技股份有限公司 | Video coding method and video coding device |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105915227A (en) * | 2016-04-08 | 2016-08-31 | 苏州大学 | Adaptive mixed data lossless compression system |
| CN105915228A (en) * | 2016-04-08 | 2016-08-31 | 苏州大学 | Adaptive mixed data lossless compression method |
| CN105915228B (en) * | 2016-04-08 | 2018-12-14 | 苏州大学 | Adaptive mixed data lossless compression method |
| CN105915227B (en) * | 2016-04-08 | 2019-03-15 | 苏州大学 | Adaptive mixed lossless date-compress system |
| CN108155969A (en) * | 2016-12-02 | 2018-06-12 | 北京深鉴科技有限公司 | A kind of decoding circuit for variable length coded data |
| CN108155969B (en) * | 2016-12-02 | 2021-02-19 | 赛灵思公司 | Decoding circuit for variable length coded data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104349165B (en) | 2017-06-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103067022B (en) | A kind of integer data lossless compression method, decompression method and device | |
| CN112399181B (en) | Image coding and decoding method, device and storage medium | |
| CN100423582C (en) | Method and device for binary coding wait coding data | |
| CN101878651B (en) | Variable length coding of coefficient clusters for image and video compression | |
| CN104349165A (en) | High-performance length-variable coding and decoding methods and devices | |
| CN103546161A (en) | Lossless compression method based on binary processing | |
| CN101841707A (en) | High-speed real-time processing arithmetic coding method based on JPEG 2000 standard | |
| CN114222129A (en) | Image compression encoding method, image compression encoding device, computer equipment and storage medium | |
| CN105409129B (en) | Encoder device, decoder apparatus and method | |
| CN104104954A (en) | Tile-based compression and decompression for graphic applications | |
| CN101878646B (en) | Coding method and device for joint amplitude and position of video compression coefficients | |
| CN103597829A (en) | Method for coding video quantization parameter and method for decoding video quantization parameter | |
| CN103428502B (en) | Decoding method and decoding system | |
| CN110191341B (en) | Depth data coding method and decoding method | |
| US8817875B2 (en) | Methods and systems to encode and decode sequences of images | |
| CN109660809A (en) | Based on the decoded colmv data lossless compression method of inter and system | |
| CN101998122A (en) | Method and device for decoding normal form Hoffman hardware in JPEG (Joint Photographic Expert Group) image | |
| CN102388538A (en) | Method and device for encoding an input bit sequence and corresponding decoding method and device | |
| CN102651795B (en) | Run-length reduced binary sequence compressed encoding method | |
| US8798139B1 (en) | Dual-pipeline CABAC encoder architecture | |
| CN102298782A (en) | System, method, and computer program product for parameter estimation for lossless video compression | |
| US9407918B2 (en) | Apparatus and method for coding image, and non-transitory computer readable medium thereof | |
| US20160323603A1 (en) | Method and apparatus for performing an arithmetic coding for data symbols | |
| CN116170599B (en) | Synchronous real-time image compression method, system, medium and terminal | |
| CN201910880U (en) | Hardware decoding device for canonical Huffman in JPEG (joint photographic experts group) image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |