CN103152567A - Columbus encoder with index of any order and method thereof - Google Patents
Columbus encoder with index of any order and method thereof Download PDFInfo
- Publication number
- CN103152567A CN103152567A CN2013100601009A CN201310060100A CN103152567A CN 103152567 A CN103152567 A CN 103152567A CN 2013100601009 A CN2013100601009 A CN 2013100601009A CN 201310060100 A CN201310060100 A CN 201310060100A CN 103152567 A CN103152567 A CN 103152567A
- Authority
- CN
- China
- Prior art keywords
- length
- binary coded
- encoded radio
- exponent number
- coded value
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 16
- 230000001186 cumulative effect Effects 0.000 claims description 11
- 238000005267 amalgamation Methods 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 4
- 230000006835 compression Effects 0.000 abstract description 4
- 238000007906 compression Methods 0.000 abstract description 4
- 238000009825 accumulation Methods 0.000 abstract description 3
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000012545 processing Methods 0.000 description 5
- 238000003860 storage Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 108020004705 Codon Proteins 0.000 description 1
- 240000001973 Ficus microcarpa Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Images
Landscapes
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
本发明公开了一种任意阶数指数哥伦布编码器及其方法，编码器至少包括编码值计算模块、编码值长度计算模块、累加控制模块和码流生成模块，各个模块内部都可采用流水操作，充分利用硬件的并行性来提高编码速度。其编码方法，只需利用加法、或、移位和比较操作就可以快速的产生指数哥伦布编码码字，适合于硬件实现。本发明用以解决视频图像编码中，由于熵编码运算复杂度高、计算量过大而引起的熵编码速度低下，难以满足超高清或高清视频实时压缩的问题。本发明可应用于超高清或高清实时视频编码芯片等对编码速度要求较高的领域中，还可以满足更高速度的视频图像编码需求。
The invention discloses an arbitrary-order exponential Columbus encoder and a method thereof. The encoder at least includes a coded value calculation module, a coded value length calculation module, an accumulation control module and a code stream generation module. Pipeline operations can be used inside each module. Take full advantage of hardware parallelism to increase encoding speed. Its encoding method can quickly generate exponential Columbus code words only by using addition, or, shift and comparison operations, and is suitable for hardware implementation. The invention is used to solve the problem of low entropy coding speed caused by high entropy coding operation complexity and excessive calculation amount in video image coding, and it is difficult to meet the problem of real-time compression of ultra-high-definition or high-definition video. The present invention can be applied to fields requiring high encoding speed, such as ultra-high-definition or high-definition real-time video encoding chips, and can also meet higher-speed video image encoding requirements.
Description
Technical field
The present invention relates to the Digital Video Processing field, relate in particular to a kind of index Columbus encoder and coding method thereof.
Background technology
Along with the development of audio frequency and video industry, people also improve day by day to the requirement of audio/video coding technology, occurred with MPEG-4, H.264/AVC waited video encoding standard, and China has also formulated Chinese video encoding standard AVS.Above-mentioned video encoding standard can under the prerequisite that keeps Subjective video quality, be compressed vision signal, thereby reduce significantly memory space and transmission bandwidth requirement efficiently.
Wherein, entropy coding is part indispensable in above-mentioned video encoding standard, can effectively remove the statistical redundancy in compression bit stream, further improves the compression efficiency of video.Entropy coding commonly used has Huffman (Huffman) coding, arithmetic coding and index Columbus coding.Huffman (Huffman) coding needs storage code table in advance, and the probabilistic model that uses fixes, and the probabilistic statistical characteristics of actual code stream is not changeless.The computation complexity of arithmetic encoder own is higher, and need to need to carry out repeatedly Bit-Plane Encoding to each data, is difficult to reach coding rate at a high speed.The index Columbus coding is a kind of variable length prefix code, code word is very regular, can calculate code word by hardware, not only need not to set up in advance and the storage code table, but also can adjust flexibly exponent number k according to the information probability distribution function, thereby reach very high code efficiency.Because it has above advantage, index Columbus coding by international video encoding standard H.264/AVC and the video encoding standard AVS employing of China as the part of entropy coding.
At present, entropy coding is generally to adopt the method for software to realize on digital signal processor (DSP, Digital Signal Processor) platform.Adopt the entropy coder of the method realization of software, speed is lower usually, is difficult to satisfy the rate request of real-time coding; In recent years, many entropy coding methods based on arithmetic coding are achieved on hardware, reach the real-time coding rate request of high definition (1920 * 1080) or SD video, but still can't obtain the real-time coding speed of ultra high-definition (3840 * 2160) video.
Summary of the invention
Technology of the present invention is dealt with problems: overcome the deficiencies in the prior art, a kind of any exponent number index Columbus encoder and coding method thereof are provided, in order to solve in encoding video pictures, computational complexity is high, the excessive entropy coding rate that causes of amount of calculation is low, is difficult to satisfy the problem of ultra high-definition or HD video Real Time Compression because entropy is encoded.
The present invention is achieved by the following technical solutions:
A kind of any exponent number index Columbus encoder comprises at least:
The encoded radio computing module utilizes sample value and the index Columbus coding exponent number inputted, calculates the binary coded value of this sample; Described binary coded value is that input sample value and 1 results added of inputting the exponent number position that moves to left is obtained;
Encoded radio length computation module, the binary coded value of utilizing the encoded radio computing module to obtain, and the index Columbus coding exponent number of input calculate the length of binary coded value;
Cumulative control module, the binary coded value length that encoded radio length computation module is obtained adds up, and produces shift control signal and code stream output control signal;
The code stream generation module, shift control signal and code stream output control signal according to cumulative control module output, use " displacement " and OR operation to complete amalgamation and the output of code stream, and the encoded radio that needs at least to use two registers to deposit respectively and obtain and the encoded radio after amalgamation.
Further, described encoded radio length computation module uses look-up table and adder to obtain the length of encoded radio.
Further, the degree of depth of depositing the register of encoded radio after encoded radio and amalgamation will equate, and will reach at least 2 times of input Sample Maximal value code length.
Another technical scheme of the present invention is as follows:
A kind of any exponent number index Columbus encoding method comprises the steps:
A) read in sample value and exponent number K;
B) calculate binary coded value and the length thereof of sample value;
C) binary coded value that obtains is buffered in a register, when the code stream length that cushions in register reaches appointment output length, the code stream of output designated length;
D) if read in the finish command, finish coding; Otherwise repeat A-D.
Further, by inputting the move to left results added of K position of sample value and 1, obtain binary coded value; This binary coded value is searched for one by one, obtain first ' 1 ' to low level from a high position apart from the distance L of lowest order, the result of calculating 2L-1+K is exactly the length of this binary coded value;
Further, the binary coded value of input is right after a sample binary coded value thereafter, leaves the high order end of a register in; After the code stream length in this register reaches appointment output length, begin to take out the encoded radio of appointment output length from the high position of this register, export as code stream; Simultaneously remaining encoded radio in register is displaced to this register highest order place.
Beneficial effect of the present invention is: (1) modules of the present invention only related to addition or, displacement and compare operation, computation complexity is low, suitable hardware is realized fast; (2) pipeline organization has all been adopted in modules inside, takes full advantage of the concurrency of hardware; (3) be easy to hardware and realize, coding rate is fast, can satisfy the real-time coding of ultra high-definition video and even higher coding requirement; (4) required hardware resource is few.
Description of drawings
Fig. 1 is the external interface of a kind of any exponent number index Columbus encoder of the present invention;
Fig. 2 is the structure chart of a kind of any exponent number index Columbus encoder of the present invention;
Fig. 3 is interface and the internal core structure thereof of encoded radio computing module of the present invention;
Fig. 4 is interface and the internal core structure thereof of encoded radio length computation module of the present invention;
Fig. 5 is interface and the internal core structure thereof of cumulative control module of the present invention;
Fig. 6 is interface and the internal core structure thereof of code stream generation module of the present invention;
Fig. 7 is that a kind of arbitrary characteristics Columbus of the present invention is coded in the coding result that field programmable logic device (FPGA) realization obtains;
Fig. 8 is a kind of arbitrary characteristics Columbus encoding method flow chart of the present invention.
Embodiment
Below in conjunction with accompanying drawing and instantiation, the present invention is described in detail.
The index Columbus coding is a kind of variable length prefix code, its hardware is realized simple, need not to set up in advance and the storage code table, not only can calculate by hardware and produce fast code word, and can adjust flexibly exponent number k according to the information probability distribution function, thereby can reach very high code efficiency.Its code word is comprised of two parts: monobasic code+group internal label.It is the k=0 of 0-9 that table 1 has been listed sample, 1,2,3 index Columbus coding result.
The 0-3 level index Columbus coding of table 1 sample 0-9
Because index Columbus coding can only be to encoding more than or equal to zero integer, to the integral sample that has negative, can carry out preliminary treatment before being input to the index Columbus encoder, negative is mapped in the positive number scope.Non-negative mapping can adopt formula (1) to carry out:
Fig. 1 is a kind of any exponent number index Columbus encoder external interface schematic diagram of the present invention.Input signal comprises: input data D, useful signal V, coding exponent number K, end signal E, clock signal clk and reset signal rst_n; Output signal comprises: output code flow CB, useful signal VCB and end signal CE.The sample value that input data D indicates to encode; Useful signal V represents when being high that these input data are effective, otherwise is invalid; The exponent number of the index Columbus coding that the sample value that represents coding exponent number K to input will adopt; When end signal E is high, simultaneously that the accumulation length in the index Columbus encoder is clear 0 even in index Columbus coding, existing code stream does not reach set output code flow length and code stream will be exported yet, end-of-encode; Clock clk represents the clock of this index Columbus encoder work; Reset signal rst_n is 0 o'clock, register and operating state clear 0 in inner all modules of index Columbus coding.Output code flow CB represents the code stream that obtains; When useful signal VCB was high, the code stream CB that expression obtains was effective, otherwise invalid.End signal CE represents that the index Columbus encoder encodes finishes, the accumulation length in the index Columbus encoder clear 0.
As shown in Figure 2, any exponent number index Columbus fgs encoder device of the present invention comprises: encoded radio computing module, encoded radio length computation module, cumulative control module and code stream generation module.The below will describe in detail to modules and workflow thereof.
See also Fig. 3 (a), the input signal of encoded radio computing module has input data D, useful signal V, coding exponent number K, clock clk and reset signal rst_n; Output signal is encoded radio C and useful signal V_C.Wherein, the input signal input signal of index Columbus coding namely; Output encoder value C represents to input the binary coded value of sample; Useful signal V_C is high, represents that this encoded radio C is effective, otherwise invalid.Fig. 3 (b) has showed the core texture of encoded radio computing module module: at first to the 1 K position that moves to left, then namely obtain encoded radio C with input data D addition.This module divides two clocks to complete, and the dotted line in figure represents that streamline cuts apart.
See also Fig. 4 (a), the input signal of encoded radio length computation module has encoded radio C, useful signal V_C, coding exponent number K_2t, clock clk and reset signal rst_n2t; Output signal has code word size L_C and useful signal V_L.Wherein, input signal encoded radio C, useful signal V_C are the output signal of encoded radio computing module; Because the encoded radio computing module needs 2 clocks, so the input exponent number K_2t of this module and the rst_n_2t that resets represent respectively the signal of K and 2 clocks of rst_n time-delay; Output codons length L _ C is the length of encoded radio C; Useful signal V_L is high, and L_C is effective in expression output, otherwise invalid.Fig. 4 (b) has showed the core texture of encoded radio length computation module: at first output encoder value C is carried out head 1 and detect, find that in encoded radio C, first from a high position to the low level ' 1 ' apart from the distance L of lowest order, then obtains the value of 2L-1.The present invention adopts the thinking of look-up table, has completed above-mentioned two steps in a clock.Table 2 has been listed the example of a look-up table, and the value of priority is larger, and expression priority is higher.This module also divides two clocks to complete, and the dotted line in figure represents that streamline cuts apart.
Look-up table example of table 2
See also Fig. 5 (a), the input signal of cumulative control module has code word size L_C, useful signal V_L, end signal E_4t, clock clk and reset signal rst_n_4t; Output signal has Len_A, Valid_L and G_then_CL.Code word size L_C and useful signal V_L are the output signals of encoded radio length computation module; End signal E_4t and reset signal rst_n_4t are respectively the signal after end signal E and 4 clocks of reset signal rst_n time-delay; Output signal Len_A represents the total length after L_C adds up; Valid_L is high, and the Len_A of expression output is effective; G_then_CL for low expression the total length L en_A after cumulative less than set output code flow length C L.Fig. 5 (b) has showed the core texture of cumulative control module: if the length of Len_A less than set code stream length C L, code word size L_C and Len_A are cumulative; If the length of Len_A more than or equal to set output code flow length C L, after Len_A deducts CL, then adds up with code word size C_L.Simultaneously G_than_CL is set high.This module divides three clocks to complete, and the dotted line in figure represents that streamline cuts apart.
See also Fig. 6 (a), the input signal of code stream generation module has Len_A, Valid_L, G_than_CL, encoded radio C_4t, clock clk and the rst_n_7t that resets; Output signal has code stream CB, useful signal VCB and end signal E_C.Wherein Len_A, Valid_L, G_than_CL are the output signal of cumulative control module, and encoded radio C_4t is the signal after 4 clocks of encoded radio C time-delay of encoded radio computing module output; The rst_n_7t that resets represents the signal of 7 clocks of reset signal rst_n time-delay.The output signal of any exponent number index Columbus of namely the present invention of output signal encoder.Fig. 6 (b) has showed the core texture of code stream generation module module: the initial value of register d1, d2, c1 and c2 in this module is all 0.At first with encoded radio C_4t assignment to register d1, then to the right cyclic shift Len_A assignment to register d2, then register d2 and register c1 by or operation realize the amalgamation of code stream, if input signal G_than_CL be high, the high CL position of register c2 is exported as code stream CB.Simultaneously the value of register c1 also needs register c2 to constantly update, if G_than_CL be height, and the value of the register c1 CL position that need to move to left; If G_than_CL is low, do not make shifting processing.This module divides three clocks to complete, and the dotted line in figure represents that streamline cuts apart.Here, the degree of depth of register d1, d2, c1 and c2 need to be greater than peaked 2 times of input coding value C_4t, carrying out peeks from register c2 all needs time-delay because the G_than_CL signal is from producing to, so need to reserve to these registers the space of at least 2 times.In figure, the degree of depth of register d1, d2, c1 and c2 is set as 128.In example, we are set as 64 with set code stream length C L.
In order to verify validity of the present invention, field programmable gate array (FPGA) programming that any exponent number index Columbus encoder of the present invention is XC5VFX70T in model realizes, the highest frequency that comprehensively obtains is 393.918MHz, used 502 register cells (Slice Registers) and 533 look-up table unit (Slice LUTs), institute's cost source is about 1% of this model FPGA total resources.Fig. 7 downloads to any exponent number index Columbus encoder sequence of the present invention the coding result that obtains after the FPGA operation of above-mentioned model, and in figure, reset_n represents to input reset signal; Level_K presentation code exponent number k; Data_in_24 is the input sample value; Valid_in is the useful signal of input number, and height is effective; Data_end_in represents the end of input signal; Code_out represents output code flow; Valid_out represents the useful signal of output code flow, and height is effective; Code_end_out represents the end of output signal.Fig. 7 (a) is continuously the integer of input 0-256, the result of using 0 rank index Columbus coding to obtain; Fig. 7 (b) is continuously the integer of input 0-256, the result of using the index Columbus coding on 1 rank to obtain; Fig. 7 (c) is for being interrupted the integer of input 0-256, the result that the index Columbus coding of use exponent number from 0 to 15 circulation change obtains; Fig. 7 (d) is continuously the integer of input 0-256, uses the index Columbus coding of exponent number from 0 to 15 circulation change, the coding result that obtains after the end signal of a clock of input.
In conjunction with flow chart shown in Figure 8, a kind of step of any exponent number index Columbus encoding method is as follows:
Binary coded value and the length thereof of step 2, calculating sample value;
If step 4 is read in the finish command, finish coding; Otherwise repeat A-D.
In step 3, by inputting the move to left results added of K position of sample value and 1, obtain binary coded value; This binary coded value is searched for one by one, obtain first ' 1 ' to low level from a high position apart from the distance L of lowest order, the result of calculating 2L-1+K is exactly the length of this binary coded value.As input that sample value is 5, exponent number is 0, can obtain binary coded value and be (101)
B+ 1=(110)
B, subscript B represents binary representation; Can get encoded radio length is 2*3-1+0=5, so the code word that obtains is (00110)
B
In step 4, the binary coded value of input is right after a sample binary coded value thereafter, leaves the high order end of a register in; After the code stream length in this register reaches appointment output length, begin to take out the encoded radio of appointment output length from the high position of this register, export as code stream; Simultaneously remaining encoded radio in register is displaced to this register highest order place.If encoder reads in sample 5,6,7,8 continuously, exponent number is all 0, and the code stream that cushions in register is 001100011100010000001001, is 16 if specify output length, the code stream of output is 0011000111000100, and in register, remaining code stream is 00001001.
Due to, the present invention is after the FPGA that model is XC5VFX70T realizes, and the highest frequency that comprehensively obtains is 393.918MHz, and namely the theoretical peak processing speed can reach per second 393.918M pixel.The real-time coding of ultra high-definition video needs per second to process the ultra high-definition image of 25-30 frame 3840 * 2160, and its maximum processing speed reaches per second 3840 * 2160 * 30=248.832M pixel.So, in actual applications, even dominant frequency of the present invention is reduced slightly, also can satisfy the requirement of ultra high-definition video real-time coding fully.Certainly, application of the present invention is not limited to this, and the present invention is embedded in very lagre scale integrated circuit (VLSIC) (VLSI), and processing speed can further promote, and is enough to satisfy rate request higher application scenario.
The non-elaborated part of the present invention belongs to those skilled in the art's known technology.
The above is only better embodiment of the present invention, and need not with the restriction the present invention.Within the spirit and principles in the present invention all, any modification of doing, the merging that is equal to replacement, functional module or expansion and improvement etc., all should be included in of the present invention comprise scope within.
Claims (6)
1. any exponent number index Columbus encoder is characterized in that comprising at least:
The encoded radio computing module utilizes sample value and the index Columbus coding exponent number inputted, calculates the binary coded value of this sample; Described binary coded value is that input sample value and 1 results added of inputting the exponent number position that moves to left is obtained;
Encoded radio length computation module, the binary coded value of utilizing the encoded radio computing module to obtain, and the index Columbus coding exponent number of input calculate the length of binary coded value;
Cumulative control module, the binary coded value length that encoded radio length computation module is obtained adds up, and produces shift control signal and code stream output control signal;
The code stream generation module, shift control signal and code stream output control signal according to cumulative control module output, use " displacement " and OR operation to complete amalgamation and the output of code stream, and the encoded radio that needs at least to use two registers to deposit respectively and obtain and the encoded radio after amalgamation.
2. any exponent number index Columbus encoder as claimed in claim 1 is characterized in that: described encoded radio length computation module uses look-up table and adder to obtain the length of encoded radio.
3. any exponent number index Columbus encoder as claimed in claim 1 is characterized in that: the deep equality of described two registers, and the degree of depth will reach 2 times of input Sample Maximal value code length at least.
4. any exponent number index Columbus encoding method, is characterized in that comprising the steps:
A) read in sample value and exponent number K;
B) calculate binary coded value and the length thereof of sample value;
C) binary coded value that obtains is buffered in a register, when the code stream length that cushions in register reaches appointment output length, the code stream of output designated length;
D) if read in the finish command, finish coding; Otherwise repeating step A-D.
5. a kind of any exponent number index Columbus encoding method as claimed in claim 4 is characterized in that: by inputting the move to left results added of K position of sample value and 1, obtain binary coded value; This binary coded value is searched for to low level one by one from a high position, obtain first " 1 " apart from the distance L of lowest order, the result of calculating 2L-1+K is exactly the length of this binary coded value;
6. a kind of any exponent number index Columbus encoding method as claimed in claim 4, it is characterized in that: the binary coded value of input is right after a sample binary coded value thereafter, leaves the high order end of a register in; After the code stream length in this register reaches appointment output length, begin to take out the encoded radio of appointment output length from the high position of this register, export as code stream; Simultaneously remaining encoded radio in register is displaced to this register highest order place.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100601009A CN103152567A (en) | 2013-02-26 | 2013-02-26 | Columbus encoder with index of any order and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100601009A CN103152567A (en) | 2013-02-26 | 2013-02-26 | Columbus encoder with index of any order and method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103152567A true CN103152567A (en) | 2013-06-12 |
Family
ID=48550409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100601009A Pending CN103152567A (en) | 2013-02-26 | 2013-02-26 | Columbus encoder with index of any order and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103152567A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110447222A (en) * | 2017-03-27 | 2019-11-12 | 索尼公司 | Image-forming component and imaging device and method |
| WO2023098803A1 (en) * | 2021-12-03 | 2023-06-08 | 维沃移动通信有限公司 | Point cloud encoding processing method, point cloud decoding processing method and related device |
| CN116505952A (en) * | 2023-06-27 | 2023-07-28 | 厦门立林科技有限公司 | Infrared code compression method, device, intelligent device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050275570A1 (en) * | 2004-06-10 | 2005-12-15 | Wang Wen-Shan Vincent | Parallel leading bit detection for Exp-Golomb decoding |
| JP2007295157A (en) * | 2006-04-24 | 2007-11-08 | Matsushita Electric Ind Co Ltd | Data encoding apparatus, data encoding method, data encoding program, and information recording medium recording the data encoding program |
| CN101304534A (en) * | 2008-06-20 | 2008-11-12 | 四川长虹电器股份有限公司 | Index Columbus encoding method |
| CN102413324A (en) * | 2010-09-20 | 2012-04-11 | 联合信源数字音视频技术（北京）有限公司 | Precoding code table optimization method and precoding method |
-
2013
- 2013-02-26 CN CN2013100601009A patent/CN103152567A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050275570A1 (en) * | 2004-06-10 | 2005-12-15 | Wang Wen-Shan Vincent | Parallel leading bit detection for Exp-Golomb decoding |
| JP2007295157A (en) * | 2006-04-24 | 2007-11-08 | Matsushita Electric Ind Co Ltd | Data encoding apparatus, data encoding method, data encoding program, and information recording medium recording the data encoding program |
| CN101304534A (en) * | 2008-06-20 | 2008-11-12 | 四川长虹电器股份有限公司 | Index Columbus encoding method |
| CN102413324A (en) * | 2010-09-20 | 2012-04-11 | 联合信源数字音视频技术（北京）有限公司 | Precoding code table optimization method and precoding method |
Non-Patent Citations (1)
| Title |
|---|
| 孙健: "高速指数哥伦布编解码器的VLSI设计与实现", 《半导体光电》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110447222A (en) * | 2017-03-27 | 2019-11-12 | 索尼公司 | Image-forming component and imaging device and method |
| WO2023098803A1 (en) * | 2021-12-03 | 2023-06-08 | 维沃移动通信有限公司 | Point cloud encoding processing method, point cloud decoding processing method and related device |
| CN116505952A (en) * | 2023-06-27 | 2023-07-28 | 厦门立林科技有限公司 | Infrared code compression method, device, intelligent device and storage medium |
| CN116505952B (en) * | 2023-06-27 | 2023-09-08 | 厦门立林科技有限公司 | Infrared code compression method, device, intelligent device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101534373B (en) | Hardware Implementation Method of Near Lossless Compression of Remote Sensing Image Based on Improved JPEG-LS Algorithm | |
| CN101466041B (en) | A task scheduling method for multi-viewpoint video coding for multi-core processors | |
| CN105120293A (en) | Image cooperative decoding method and apparatus based on CPU and GPU | |
| CN101404772A (en) | VLSI image compression encoder based on wavelet transformation | |
| CN101867809A (en) | High-speed image compression VLSI coding method and coder based on systolic array | |
| CN101917622A (en) | 14bit wide image compression hardware coder | |
| CN115412735A (en) | Encoding and decoding method, apparatus, device, storage medium, and computer program | |
| CN109391816B (en) | Parallel processing method of entropy coding in HEVC based on CPU+GPU heterogeneous platform | |
| CN103152567A (en) | Columbus encoder with index of any order and method thereof | |
| CN117692662A (en) | A point cloud encoding and decoding method based on double octree structure | |
| CN102724499A (en) | Variable-compression ratio image compression system and method based on FPGA | |
| CN102833546A (en) | High-speed image compression method and device based on wavelet sub-band interleaving optimal quantization | |
| CN102547291B (en) | Field programmable gate array (FPGA)-based joint photographic experts group (JPEG) 2000 image decoding device and method | |
| CN103200407A (en) | An Adaptive Entropy Encoder | |
| CN102223534B (en) | All-parallel bit plane coding method for image compression | |
| CN102333222B (en) | Two-dimensional discrete wavelet transform circuit and image compression method using same | |
| CN102523455B (en) | Multi-thread arithmetic coding circuit and method based on standard JPEG 2000 | |
| Li et al. | A low power, fully pipelined JPEG-LS encoder for lossless image compression | |
| CN101998122A (en) | Method and device for decoding normal form Hoffman hardware in JPEG (Joint Photographic Expert Group) image | |
| CN111479286A (en) | A data processing method for edge computing system to reduce communication traffic | |
| CN103067718B (en) | Be applicable to the one-dimensional discrete cosine inverse transform module circuit of digital video decoding | |
| CN106210725A (en) | A kind of video image compression perception coder/decoder system and decoding method thereof | |
| CN1207917C (en) | High-speed low power consumption MQ encoder applicable dto JPEG2000 standard | |
| CN101854538B (en) | Motion image processing method and motion image processor | |
| CN117197262A (en) | Semantic scalable image coding method, system, device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130612 |