CN102547281A - Joint photographic experts group (JPEG) image compression concurrency control method based on digital signal processor (DSP) - Google Patents

Joint photographic experts group (JPEG) image compression concurrency control method based on digital signal processor (DSP) Download PDF

Info

Publication number
CN102547281A
CN102547281A CN2011104510368A CN201110451036A CN102547281A CN 102547281 A CN102547281 A CN 102547281A CN 2011104510368 A CN2011104510368 A CN 2011104510368A CN 201110451036 A CN201110451036 A CN 201110451036A CN 102547281 A CN102547281 A CN 102547281A
Authority
CN
China
Prior art keywords
dsp
jpeg
memory
data
block
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
Application number
CN2011104510368A
Other languages
Chinese (zh)
Inventor
王玉周
刘天宇
董铁军
付存伟
宋晓航
李丹
于维双
赵全邦
袁锟
孙利涛
张雷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SHENYANG JUDE VIDEO TECHNOLOGY Co Ltd
Original Assignee
SHENYANG JUDE VIDEO TECHNOLOGY Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by SHENYANG JUDE VIDEO TECHNOLOGY Co Ltd filed Critical SHENYANG JUDE VIDEO TECHNOLOGY Co Ltd
Priority to CN2011104510368A priority Critical patent/CN102547281A/en
Publication of CN102547281A publication Critical patent/CN102547281A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention relates to a joint photographic experts group (JPEG) image compression concurrency control method based on a digital signal processor (DSP), which comprises the following steps: adjusting a JPEG algorithm structure, adopting a step-by-step completed strategy for image data partitioning operation, namely encoding while partitioning, and simultaneously transferring partitioned data block units into an internal memory to be processed; opening up two memory areas in the internal memory; dividing JPEG algorithm by combination with encoding structure characteristics and a hardware structure of the DSP, achieving concurrency completion in the two memory areas in the internal memory through the DSP and a peripheral device enhanced direct memory access (EDMA), and achieving concurrent execution of the JPEG algorithm on the DSP based on the concurrency completion. The JPEG image compression concurrency control method prevents a processor from reading and writing an external memory, greatly prolongs JPEG encoding time, provides conditions for JPEG real-time processing, and improves utilization rate and algorithm execution efficiency of the DSP.

Description

Jpeg image compression concurrency control method based on DSP
Technical field
The present invention relates to the Image Compression in a kind of digital information technology, a kind of specifically jpeg image compression concurrency control method based on DSP.
Background technology
JPEG (Joint Photographic Experts Group) is the committee that under the International Organization for Standardization leader, formulates the still image compression standard, and the first international still image compression standard ISO 10918-1 of cover (JPEG) is exactly that this committee formulates.Because the good quality of JPEG makes ItsObtain great success in recent years, be widely used in the Internet and digital camera field.Jpeg format is a most popular picture format on the present network, is can be File Compress to minimum form, and 80% image has all adopted the JPEG compression standard on the website.Yet JPEG compression algorithm complexity is high, coding rate is slow, and is higher to the hardware handles Capability Requirement that algorithm is realized.
DSP (digital signal processor) is a kind of microprocessor of uniqueness; It is the device of handling bulk information with digital signal; Having data-handling capacity and programmability at a high speed, is the embedded processing chip that becomes more and more important in the digitalized electron world.The powerful data-handling capacity of DSP and the high speed of service not only are embodied on the kernel processes ability, more are embodied in its peripheral hardware interface with high integration and coprocessor EDMA etc., have guaranteed the concurrency of DSP data processing and peripheral unit control.
DSP at first will handle each piece the view data piecemeal carrying out jpeg image when compression afterwards, is discrete cosine transform, quantification, Z-transformation successively, encodes, is combined into code stream, jpeg coder as shown in Figure 1.After deblocking was accomplished, except cataloged procedure has the continuity requirement to data block, all the other operations all were independently to the processing of each piece.So when carrying out the JPEG compression:
1. externally create in the memory one with the onesize storage area preservation image block of image after data;
2. each blocks of data unit is carried out discrete cosine transform, quantification, Z-transformation, encodes, is combined into code stream.
Because image data amount is big, when carrying out deblocking, has destroyed the continuity of view data, so need externally to open a memory block in the memory again the view data behind the preservation piecemeal.Carrying out the moving in the process of view data, be simultaneously external memory storage to be carried out read-write operation, must influence the execution efficient of DSP.
When carrying out subsequent operation, action needs such as discrete cosine transform, quantification, coding carry out lot of data to be handled, and can data block be moved in the internal memory and operate.After accomplishing the moving of data block, it is carried out subsequent operation such as discrete cosine transform after, the data code flow that generates is write in the external memory storage, carry out moving of next data block afterwards, accomplish the encoding operation of each data block successively.Will visit in a large number external memory storage like this, and the sequential processes of data, have a strong impact on the concurrency that program is carried out, processor and peripheral hardware thereof are not effectively utilized, influenced the operating efficiency of DSP.
Summary of the invention
The sequential processes of data has a strong impact on the concurrency that program is carried out when in the prior art data being compressed; Influence the weak points such as operating efficiency of DSP, but the technical problem that the present invention will solve provides a kind of JPEG code optimization method based on DSP of practicality and high efficiency parallel control.
For solving the problems of the technologies described above, the technical scheme that the present invention adopts is:
The present invention is based on the jpeg image compression concurrency control method of DSP, it is characterized in that may further comprise the steps:
Adjustment jpeg algorithm structure, view data divides block operations to adopt the strategy of progressively accomplishing, and promptly limit piecemeal limit coding is moved the data block unit behind the piecemeal in the internal memory simultaneously and is handled;
In internal memory, open up two memory blocks;
In conjunction with the coding structure characteristics of JPEG and the hardware configuration of DSP, jpeg algorithm is divided, walk abreast by DSP and ancillary equipment EDMA thereof in two memory blocks of internal memory and accomplish, realize the executed in parallel of jpeg algorithm on DSP on this basis.
In internal memory, opening up two memory blocks is: two memory blocks are set in internal memory, i.e. and first memory block and second memory block, and utilize EDMA to set up ping-pong structure.
Jpeg algorithm in the executed in parallel on the DSP is: when DSP accomplishes the data processing to first memory block; EDMA has accomplished the data-moving of second video data block in second memory block; DSP is provided with EDMA another area data of image is moved in first memory block; DSP changes the view data of handling in second memory block simultaneously, and so circulation is all finished dealing with until the image block partial data.
The JPEG coding generates the JPEG compressed bit stream through deblocking, discrete cosine transform, quantification, Z-transformation, Huffman encoding and code stream combination.
The present invention has following beneficial effect and advantage:
1. the present invention utilizes the EDMA hardware of DSP, and with the jpeg algorithm separated into two parts, a part is accomplished by EDMA; A part is accomplished by the CPU of DSP; And make two parts executed in parallel, avoided processor read-write external memory storage, improved the JPEG scramble time greatly; For JPEG handles the condition that provides in real time, improve utilance and the algorithm of DSP and carry out efficient.
Description of drawings
Fig. 1 is the algorithm flow of jpeg coder;
Fig. 2 is the ping-pong structure figure of EDMA;
Fig. 3 realizes the flow chart of JPEG method for the present invention.
Embodiment
Below in conjunction with Figure of description the present invention is done further elaboration.
The jpeg image compression concurrency control method that the present invention is based on DSP may further comprise the steps:
Adjustment jpeg algorithm structure, view data divides block operations to adopt the strategy of progressively accomplishing, and promptly limit piecemeal limit coding is moved the data block unit behind the piecemeal in the internal memory simultaneously and is handled;
In internal memory, open up two memory blocks;
In conjunction with the coding structure characteristics of JPEG and the hardware configuration of DSP, jpeg algorithm is divided, walk abreast by DSP and ancillary equipment EDMA thereof in two memory blocks of internal memory and accomplish, realize the executed in parallel of jpeg algorithm on DSP on this basis.
In internal memory, opening up two memory blocks is: two memory blocks are set in internal memory, i.e. and first memory block and second memory block, and utilize EDMA to set up ping-pong structure.
Jpeg algorithm in the executed in parallel on the DSP is: when DSP accomplishes the data processing to first memory block; EDMA has accomplished the data-moving of second video data block in second memory block; DSP is provided with EDMA another area data (being equivalent to the 3rd video data block) of image-region is moved in first memory block; DSP changes the view data of handling in second memory block simultaneously, and so circulation is all finished dealing with until the image block partial data.
The present invention partly divides jpeg algorithm out with image block, is accomplished by EDMA, and utilizes EDMA to set up a ping-pong structure, and is as shown in Figure 2.
To internal memory, let processor do subsequent treatment the data-moving of piecemeal.At first in internal memory, define two 8 * 8 internal storage location A and B (i.e. first memory block and second memory block), be used to store 8 * 8 data cells of JPEG processed compressed, build ping-pong structure with EDMA.Initialization EDMA, with the original position of the raw address pointed image memory block of EDMA, destination address is set to the initial address of internal storage location A; Each unit of EDMA is set to a byte; Every frame is set to 8 unit, just in time is 8 pixels of delegation, and every is set to 8 frames; Generated 8 * 8 data blocks of 8 row, 8 row, EDMA just can extract from the image memory block and meet 8 * 8 required data blocks of JPEG coding like this.
After system obtains piece image, start EDMA, with 8 row, 8 row in the image upper left corner totally 64 data move among the internal storage location A.After EDMA accomplishes data-moving; Also accomplished the extraction operation of a data block; Processor is provided with EDMA with view data moving in internal storage location B afterwards; Processor carries out discrete cosine transform (being dct transform), quantification, Z-transformation, Huffman encoding, code stream combination operation to the data among the internal storage location A simultaneously, and after processor was accomplished the data processing to internal storage location A, EDMA can accomplish the data-moving of second video data block in internal storage location B fully; Processor is provided with EDMA view data is moved in internal storage location A, the view data among the processor processes internal storage location B.The realization flow of jpeg algorithm in DSP is as shown in Figure 3.The deblocking of jpeg algorithm and next code separated into two parts executed in parallel have avoided processor to visit external memory repeatedly, have improved the utilance of processor.
More than the jpeg image compression concurrency control method based on DSP provided by the present invention has been carried out detailed introduction; Principle of the present invention and embodiment are set forth; The principle of whole algorithm design has been explained in above execution mode explanation on the whole, helps to help to understand method of the present invention and core concept; Simultaneously, the those skilled in the art for the DSP exploitation can on execution mode and range of application, all have certain change part to its algorithm design and hardware arrangement according to thought of the present invention.

Claims (4)

1. the jpeg image based on DSP compresses concurrency control method, it is characterized in that may further comprise the steps:
Adjustment jpeg algorithm structure, view data divides block operations to adopt the strategy of progressively accomplishing, and promptly limit piecemeal limit coding is moved the data block unit behind the piecemeal in the internal memory simultaneously and is handled;
In internal memory, open up two memory blocks;
In conjunction with the coding structure characteristics of JPEG and the hardware configuration of DSP, jpeg algorithm is divided, walk abreast by DSP and ancillary equipment EDMA thereof in two memory blocks of internal memory and accomplish, realize the executed in parallel of jpeg algorithm on DSP on this basis.
2. by the described jpeg image compression concurrency control method of claim 1 based on DSP; It is characterized in that: in internal memory, opening up two memory blocks is: two memory blocks are set in internal memory; I.e. first memory block and second memory block, and utilize EDMA to set up ping-pong structure.
3. by the described jpeg image compression concurrency control method of claim 1 based on DSP; It is characterized in that: jpeg algorithm in the executed in parallel on the DSP is: when DSP accomplishes the data processing to first memory block; EDMA has accomplished the data-moving of second video data block in second memory block; DSP is provided with EDMA another area data of image is moved in first memory block; DSP changes the view data of handling in second memory block simultaneously, and so circulation is all finished dealing with until the image block partial data.
4. by the described jpeg image compression concurrency control method based on DSP of claim 1, it is characterized in that: the JPEG coding generates the JPEG compressed bit stream through deblocking, discrete cosine transform, quantification, Z-transformation, Huffman encoding and code stream combination.
CN2011104510368A 2011-12-29 2011-12-29 Joint photographic experts group (JPEG) image compression concurrency control method based on digital signal processor (DSP) Pending CN102547281A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2011104510368A CN102547281A (en) 2011-12-29 2011-12-29 Joint photographic experts group (JPEG) image compression concurrency control method based on digital signal processor (DSP)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2011104510368A CN102547281A (en) 2011-12-29 2011-12-29 Joint photographic experts group (JPEG) image compression concurrency control method based on digital signal processor (DSP)

Publications (1)

Publication Number Publication Date
CN102547281A true CN102547281A (en) 2012-07-04

Family

ID=46353081

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2011104510368A Pending CN102547281A (en) 2011-12-29 2011-12-29 Joint photographic experts group (JPEG) image compression concurrency control method based on digital signal processor (DSP)

Country Status (1)

Country Link
CN (1) CN102547281A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102801981A (en) * 2012-08-30 2012-11-28 西安空间无线电技术研究所 Multipath compressed kernel parallel encoding control method on basis of JPEG-LS (Joint Pho-tographic Experts Group-Lossless Standard) algorithm
CN103678204A (en) * 2013-12-30 2014-03-26 龙芯中科技术有限公司 Processor and data processing method
CN111274996A (en) * 2020-02-14 2020-06-12 深圳英飞拓智能技术有限公司 Face picture feature comparison method and device, computer equipment and storage medium
CN114428595A (en) * 2022-01-28 2022-05-03 深圳市爱协生科技有限公司 Image processing method, image processing device, computer equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1595445A (en) * 2004-07-01 2005-03-16 中国科学院上海技术物理研究所 Ultra-spectrum image real-time compression system based on noise masking
CN101365138A (en) * 2008-10-10 2009-02-11 中国航天科技集团公司第五研究院第五〇四研究所 JPEG2000 image compression processing system
CN101616327A (en) * 2009-07-27 2009-12-30 无锡睿网科技有限公司 The method that multi-DSP core framework and fast parallel vision signal are handled

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1595445A (en) * 2004-07-01 2005-03-16 中国科学院上海技术物理研究所 Ultra-spectrum image real-time compression system based on noise masking
CN101365138A (en) * 2008-10-10 2009-02-11 中国航天科技集团公司第五研究院第五〇四研究所 JPEG2000 image compression processing system
CN101616327A (en) * 2009-07-27 2009-12-30 无锡睿网科技有限公司 The method that multi-DSP core framework and fast parallel vision signal are handled

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102801981A (en) * 2012-08-30 2012-11-28 西安空间无线电技术研究所 Multipath compressed kernel parallel encoding control method on basis of JPEG-LS (Joint Pho-tographic Experts Group-Lossless Standard) algorithm
CN102801981B (en) * 2012-08-30 2015-04-22 西安空间无线电技术研究所 Multipath compressed kernel parallel encoding control method on basis of JPEG-LS (Joint Pho-tographic Experts Group-Lossless Standard) algorithm
CN103678204A (en) * 2013-12-30 2014-03-26 龙芯中科技术有限公司 Processor and data processing method
CN103678204B (en) * 2013-12-30 2017-02-15 龙芯中科技术有限公司 Processor and data processing method
CN111274996A (en) * 2020-02-14 2020-06-12 深圳英飞拓智能技术有限公司 Face picture feature comparison method and device, computer equipment and storage medium
CN111274996B (en) * 2020-02-14 2023-06-09 深圳英飞拓仁用信息有限公司 Face picture feature comparison method, device, computer equipment and storage medium
CN114428595A (en) * 2022-01-28 2022-05-03 深圳市爱协生科技有限公司 Image processing method, image processing device, computer equipment and storage medium
CN114428595B (en) * 2022-01-28 2024-05-31 深圳市爱协生科技股份有限公司 Image processing method, device, computer equipment and storage medium

Similar Documents

Publication Publication Date Title
TWI339074B (en) Operating method and device of coded block flag parameter
CN103188495B (en) The method of coded video data
CN103609117B (en) Code and decode the method and device of image
CN103888777B (en) Moving picture compression decompressing device
CN103997650B (en) The method and Video Decoder of a kind of video decoding
US9948941B2 (en) Circuit, method and video decoder for video decoding
CN105472389B (en) Compression method is cached outside a kind of piece for ultra high-definition processing system for video
CN110691250B (en) Image compression apparatus combining block matching and string matching
CN103141092B (en) The method and apparatus carrying out encoded video signal for the super-resolution based on example of video compress use motion compensation
CN102547281A (en) Joint photographic experts group (JPEG) image compression concurrency control method based on digital signal processor (DSP)
CN102724499B (en) Variable-compression ratio image compression system and method based on FPGA
CN104980682A (en) Intelligent dynamic high-definition video transmission system
CN107241597B (en) A kind of reversible information hidden method of combination quaternary tree adaptive coding
US20210233280A1 (en) Encoding device control method and device, and storage medium
WO2024098821A1 (en) Av1 filtering method and apparatus
TWI473040B (en) A second deblocker in a decoding pipeline
WO2008038513A1 (en) Decoding device, decoding method, decoding program, and integrated circuit
US8045021B2 (en) Memory organizational scheme and controller architecture for image and video processing
Jiang et al. Highly paralleled low-cost embedded HEVC video encoder on TI KeyStone multicore DSP
CN104113759A (en) Video system and method and device for buffering and recompressing/decompressing video frames
CN115695806A (en) High-efficiency low-delay HEVC encoder data caching and processing method
CN101287126A (en) Rapid reading method of motion compensating data based on H.264 standard
CN109040755A (en) A kind of image pretreating device suitable for Video coding
CN112468826B (en) VVC loop filtering method and system based on multilayer GAN
CN105007490B (en) Jpeg compression algorithms based on OmapL138 chips

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20120704