CN101018327A - Discrete cosine conversion integration module and its computing combination method - Google Patents
Discrete cosine conversion integration module and its computing combination method Download PDFInfo
- Publication number
- CN101018327A CN101018327A CN 200610035021 CN200610035021A CN101018327A CN 101018327 A CN101018327 A CN 101018327A CN 200610035021 CN200610035021 CN 200610035021 CN 200610035021 A CN200610035021 A CN 200610035021A CN 101018327 A CN101018327 A CN 101018327A
- Authority
- CN
- China
- Prior art keywords
- discrete cosine
- quantization
- cosine transform
- image data
- quantification
- 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
Images
Abstract
The related discrete cosine transformation (DCT) integrated module comprises: an I/O device, a controller, a storage device for quantizing/inverse-quantizing coefficient list, an image data storage device with two storage areas, a quantizing/reverse-quantizing device, a discrete cosine transformation/inverse-transformation device, and a selector operated by the controller. This invention realizes four type of calculation in one circuit module, saves resource, and supports concurrent working for I/O and calculation to increase data process speed.
Description
Technical field
The present invention relates to the encoding and decoding field of image and video, relate in particular to a kind of discrete cosine conversion integration module, and the computing combination method of this discrete cosine conversion integration module.
Background technology
To JPEG (Join Photographic Experts Group, JPEG (joint photographic experts group)) image and MPEG (the Moving Picture Experts Group of form such as, Motion Picture Experts Group) etc. when the video of form carries out encoding and decoding, general employing discrete cosine transform and inverse discrete cosine transform are as one of its main compression algorithm, and, in the major applications, the discrete cosine transform meeting relates to quantizing process, and the inverse discrete cosine transform meeting relates to the re-quantization process.Yet, existing coding/decoding system can not be finished discrete cosine transform and quantization operations, inverse discrete cosine transform and re-quantization computing in same module, having under the situation of such application demand, the a plurality of modules of general use are finished respectively, owing to independently do not share circuit resource between the module, make coding-decoding circuit comparatively complicated, chip area increases, and cost raises.In addition, in existing coding/decoding system, above-mentioned four kinds of computings are finished by software, have taken central processing unit more operation time, have reduced the speed of data processing; Moreover, the input/output interface of existing discrete cosine transform and discrete cosine inverse transform module adopts fifo registers to come the input and output data, in doing discrete cosine transform or inverse transformation process, need synchronous input and output data, otherwise can cause the pause of conversion and inverse transformation, influence the speed of data processing equally.
In sum, in order to save circuit resource, improve data processing speed, provide a kind of and can finish discrete cosine transform simultaneously, inverse discrete cosine transform, quantize, the discrete cosine conversion integration module of four kinds of computings of re-quantization and computing combination thereof is real to be necessary, what be worth emphasizing is, discrete cosine conversion integration module of the present invention does not also mean that it can only finish the discrete cosine transform computing, on the contrary, discrete cosine conversion integration module of the present invention can be finished discrete cosine transform, inverse discrete cosine transform, quantize, four kinds of computings of re-quantization and computing combination thereof, because its core is that (inverse discrete cosine transform and discrete cosine transform are reciprocal processes in discrete cosine transform, and quantization operations is followed discrete cosine transform, the re-quantization computing is to follow inverse discrete cosine transform), therefore, expression for simplicity is referred to as discrete cosine conversion integration module with it.
Summary of the invention
Main purpose of the present invention provides a kind of discrete cosine conversion integration module of saving circuit resource, improving the integrated level of system, and it can finish discrete cosine transform, inverse discrete cosine transform, quantification, four kinds of computings of re-quantization and computing combination thereof;
An also purpose of the present invention provides a kind of discrete cosine conversion integration module that improves data processing speed;
Another main purpose of the present invention provides a kind of computing combination method of discrete cosine conversion integration module, and it can finish discrete cosine transform, inverse discrete cosine transform, quantification, four kinds of computings of re-quantization and computing combination thereof.
For achieving the above object, discrete cosine conversion integration module provided by the invention comprises input/output device, control device, quantification/re-quantization coefficient table storage device, image data storage apparatus, quantification/re-quantization device, discrete cosine transform/inverter and three selectors.Wherein, image data storage apparatus comprises two storage areas.
The computing combination method of discrete cosine conversion integration module provided by the present invention may further comprise the steps: the first step, and control device starts computing, reads the data that need conversion from image data storage apparatus; Second step, judge whether to carry out inverse discrete cosine transform: if do not carry out inverse discrete cosine transform, then enter data into and carry out discrete cosine transform in discrete cosine transform/inverter, after conversion finishes, judge whether to carry out quantization operations, if do not quantize, then directly result data is deposited in the image data storage apparatus, carry out quantization operations if desired, then quantification/re-quantization device carries out quantization operation according to the control information of control device, and then result data is deposited in the image data storage apparatus; Carry out inverse discrete cosine transform if desired, then further judge whether to carry out the re-quantization computing, if do not need to carry out the re-quantization computing, then carry out inverse discrete cosine transform by discrete cosine transform/inverter, and the inverse transformation result data deposited in the image data storage apparatus, carry out the re-quantization computing if desired, quantification/re-quantization device carries out the re-quantization operation according to the control information of control device, deposits result data in image data storage apparatus then.
Discrete cosine conversion integration module provided by the present invention with and the combinatorial operation method by control device three selectors are controlled, in a circuit module, realized discrete cosine transform, inverse discrete cosine transform, quantize, four kinds of computings of re-quantization and computing combination thereof, saved circuit resource, wherein, discrete cosine transform and quantize can concurrent operation, inverse discrete cosine transform and re-quantization also can concurrent operations, simultaneously, because image data storage apparatus is divided into two storage areas, support the input and the work of computing independent parallel of view data, improved data processing speed.
Description of drawings
Fig. 1 is that discrete cosine conversion integration module of the present invention is applied to the electrical block diagram in the coding/decoding system;
Fig. 2 is the workflow diagram of discrete cosine conversion integration module of the present invention;
Fig. 3 is the electrical block diagram that replaces selector among Fig. 1 with selector unit.
Be described in further detail below in conjunction with embodiment and accompanying drawing thereof.
Embodiment
Referring to shown in Figure 1, Fig. 1 has disclosed the circuit structure in the coding/decoding system that discrete cosine conversion integration module 1 of the present invention is applied to comprise central processing unit 2.Discrete cosine conversion integration module 1 of the present invention (part in the double dot dash line frame) comprises input/output device 100, control device 110, quantification/re-quantization coefficient table storage device 120, image data storage apparatus 130, quantification/re-quantization device 140, discrete cosine transform/inverter 160 and three selectors 150,151,152.Image data storage apparatus 130 comprise two can be by the storage area 131,132 of concurrent access, two storage areas 131,132 can be transfused to respectively/output device 100, discrete cosine transform/inverter 160 or quantification/re-quantization device visit 140 visits.
Input/output device 100 is connected with control device 110, quantification/re-quantization coefficient table storage device 120, image data storage apparatus 130 respectively, and quantification/re-quantization device 140 is connected with quantification/re-quantization coefficient table storage device 120.Be connected to each other by three selectors 150,151,152 between image data storage apparatus 130, quantification/re-quantization device 140 and the discrete cosine transform/inverter 160, concrete connected mode is introduced hereinafter in detail.
Central processing unit 2 can be by input/output device 100 to control device 110 write control informations or read state information, also can also write coefficient table by input/output device 100 vectorizations/re-quantization coefficient table storage device 120 by the data of input/output device 100 after image data storage apparatus 130 writes the data that need conversion or reads conversion.
Quantification/re-quantization device 140 and discrete cosine transform/inverter 160 can take out data from storage area 132 or 131, and change under the control of control device 110, and the result after the conversion deposits in storage area 132 or 131.Meanwhile, the result data after central processing unit 2 can write the data that need conversion and read conversion storage area 131 or 132.Finish and central processing unit visit end in data conversion, can carry out the conversion of next group data then by the mapping relations of two storage areas 131,132 of control device 110 exchanges.The stagnation that the working method of the storage area 131,132 in the discrete cosine conversion integration module 1 of the present invention has avoided available technology adopting fifo registers input and output data to be caused.
Input/output device 100 receives from central processing unit 2 and quantizes that coefficient table deposits quantifications/re-quantization coefficient table storage device 120 in, the reception view data deposits image data storage apparatus 130 (storage area 131 or storage area 132) in, and receiving control information deposits control device 110 in.
Element in the frame of broken lines 3 among Fig. 1 is that controlled device 110 is controlled, mapping, control quantification/re-quantization device 140 that control device 110 carries out storage area 131 and 132 according to the control information of central processing unit 2 are selected quantification/re-quantization mode, control discrete cosine transform/inverter 160 is selected alternative types, and control selector 150,151 and 152 is selected data flow.
Selector 150,151,152 includes two inputs and an output, in the present embodiment, interconnect between each selector and image data storage apparatus 130, quantification/re-quantization device 140, the discrete cosine transform/inverter 160, concrete annexation is as follows: the output of selector 150 connects image data storage apparatus 130, and two input connects quantification/re-quantization device 140 and discrete cosine transform/inverter 160 respectively; The output of selector 151 connects quantification/re-quantization device 140, and two input connects image data storage apparatus 130 and discrete cosine transform/inverter 160 respectively; The output of selector 152 links discrete cosine transform/inverter 160, and two input connects image data storage apparatus 130 and quantification/re-quantization device 140 respectively.
Selector 150,151 and 152 can carry out " gating " respectively to the data flow between image data storage apparatus 130, quantification/re-quantization device 140 and the discrete cosine transform/inverter 160, thereby finish the discrete cosine transform of view data, inverse discrete cosine transform quantizes or the re-quantization computing.
By description respectively to above-mentioned each main element, the course of work of discrete cosine conversion integration module of the present invention can roughly be described below: control device 110 is handled 2 instruction startup conversion according to central authorities, reading of data from image data storage apparatus 130, and control selector 150,151 and 152 are input to quantification/re-quantization device 140 to data and discrete cosine transform/inverter 160 is handled, discrete cosine transform/inverter 160 selects corresponding mapping mode to carry out conversion according to the control information of control device 110, whether quantification/re-quantization device 140 is also selected to quantize or the re-quantization computing according to the control information of control device 110, quantize if desired or the re-quantization computing, then select in corresponding quantization/re-quantization mode and quantification/re-quantization coefficient table storage device 120 corresponding coefficient table to handle, then calculated result is deposited in the image data storage apparatus 130.After corresponding computing finished, control device 110 was notified central processing unit 2 by sending the mode of interrupting or the end mark position being set, and central processing unit 2 is read transformation results by input/output device 100 from image data storage apparatus 130.
Invent described discrete cosine conversion integration module 1 and can finish discrete cosine transform, inverse discrete cosine transform, quantification, four kinds of computings of re-quantization and computing combination, the concrete working method of four kinds of main computing combinations is described below respectively:
Discrete cosine transform: in the process of some coding, need not to carry out quantization operations, can use selector 151,152 will quantize/140 bypasses of re-quantization device, only realize discrete cosine transform function if only need carry out discrete cosine transform.Detailed process is as follows: receive view data and the control information that need carry out discrete cosine transform by input/output device 100 from central processing unit 2.Control device 110 is according to control information control selector 152, select image data storage apparatus 130 (quantification/re-quantization device 140 is in bypass condition at this moment) by selector 152, and the data in the image data storage apparatus 130 are input in discrete cosine transform/inverter 160, carry out discrete cosine transform; Finish after the discrete cosine transform, control device 110 control selectors 150 are selected discrete cosine transform/inverter 160, and directly deposit transformation results wherein in image data storage apparatus 130.
Inverse discrete cosine transform computing: in some special application, need not to carry out the re-quantization computing if only inverse discrete cosine transform need be carried out, can use selector 151,152 will quantize/140 bypasses of re-quantization device, only realize the inverse discrete cosine transform function.Detailed process is as follows: receive view data and the control information that need carry out inverse discrete cosine transform by input/output device 100 from central processing unit 2.Control device 110 is according to control information control selector 152, select image data storage apparatus 130 (quantification/re-quantization device 140 is in bypass condition at this moment) by selector 152, and the data in the image data storage apparatus 130 are input in discrete cosine transform/inverter 160, carry out inverse discrete cosine transform; Finish after the discrete cosine transform, control device 110 control selectors 150 are selected discrete cosine transform/inverter 160, and directly deposit transformation results wherein in image data storage apparatus 130.
Discrete cosine transform and quantification: in the actual coding process, generally after executing the discrete cosine transform computing, all need to do quantization operations.Discrete cosine conversion integration module 1 of the present invention can be realized discrete cosine transform simultaneously and be quantized two kinds of computings by parallel working method in cataloged procedure, improved code efficiency.Detailed process is as follows: receive quantification coefficient table, view data and control information by input/output device 100 from central processing unit 2, control device 110 is according to control information control selector 152, and the data of selection image data storage apparatus 130 are input to carries out discrete cosine transform in discrete cosine transform/inverter 160; Finish after the discrete cosine transform, control device 110 control selectors 151 are input to quantification/re-quantization device 140 with the result of discrete cosine transform and carry out quantization operations; After quantization operations finished, control device 110 control selectors 150 deposited the quantized result of quantification/re-quantization device 140 in image data storage apparatus 130.
Re-quantization and inverse discrete cosine transform: in the reality decoding, all need to do earlier the re-quantization computing before the general execution inverse discrete cosine transform computing, discrete cosine conversion integration module 1 of the present invention mode by concurrent working in cataloged procedure realizes re-quantization computing and inverse discrete cosine transform computing, improves decoding efficiency greatly.Detailed process is as follows: receive quantification coefficient table, view data and control information by input/output device 100 from central processing unit 2.The data of control device 110 control selectors 151 selection image data storage apparatus 130 are input to quantification/re-quantization device 140 and carry out re-quantization; After finishing the re-quantization computing, control device 110 is controlled selectors 152, the re-quantization operation result is input to carries out the inverse discrete cosine transform computing in discrete cosine transform/inverter 160; After computing finished, control device 110 control selectors 150 deposited final transformation results in image data storage apparatus 130.
More than describe at Fig. 1 and disclosed the process that discrete cosine conversion integration module 1 of the present invention is realized four kinds of computings and computing combination respectively respectively.In fact, in most encoding-decoding process, above-mentioned four kinds of computings are not independently to carry out respectively.Flow chart among Fig. 2 organically combines four kinds of computings, clear, complete description the computing combination method of discrete cosine conversion integration module 1 of the present invention, specifically describe as follows:
The first step, control device 110 starts computing in step 40, and execution in step 41 then, read the data that need conversion from image data storage apparatus 130.
Second step, in step 42, judge whether to carry out inverse discrete cosine transform, if the judged result of step 42 is " N ", then execution in step 43, enter data into and carry out discrete cosine transform in discrete cosine transform/inverter 160, after conversion finishes, in step 431, judge whether to carry out quantization operations, if the judged result of step 431 is " N ", then direct execution in step 45, result data is deposited in the image data storage apparatus 130, if the judged result of step 431 is " Y ", then first execution in step 432, quantification/re-quantization device 140 carries out quantization operation according to the control information of control device 110, and then execution in step 45, result data is deposited in the image data storage apparatus 130.
If the judged result of step 42 is " Y ", then execution in step 44, judge whether to carry out the re-quantization computing, if the judged result of step 44 is " N ", then direct execution in step 442, discrete cosine transform/inverter 160 carries out inverse discrete cosine transform, if the judged result of step 44 is " Y ", then first execution in step 441, quantification/re-quantization device 140 carries out the re-quantization operation according to the control information of control device 110, and then execution in step 442.After the execution of step 442, execution in step 45 deposits result data in image data storage apparatus 130.
In the 3rd step, execution in step 46 judges whether total transform block number is finished (because data are that piecemeal carries out various computings, image data storage apparatus 130 can be stored 1～12 blocks of data), if the judged result of step 46 is " N ", then return execution in step 41, repeat above-mentioned steps; If the judged result of step 46 is " Y ", then execution in step 47, send interrupt signal or set end mark position by control device 110, finish computing then in step 48.It is emphasized that the discrete cosine transform/inverse transformation in the said process or the execution in step of quantification/re-quantization finish by three selectors 150,151,152.
Be appreciated that by above description, discrete cosine conversion integration module 1 of the present invention is by 110 pairs of selectors 150 of control device, 151,152 control, in a circuit module, realized discrete cosine transform, inverse discrete cosine transform, quantize, four kinds of computings of re-quantization, saved circuit resource, also improved the integrated level of system, simultaneously, because image data storage apparatus 130 is divided into two storage areas 131,132, under the control of control device 110, when input/output device 100 one of them storage area of visit, another storage area can be by discrete cosine transform/inverter 160 or 140 visits of quantification/re-quantization device, support the mode of concurrent working, mode with existing employing fifo registers is compared, and the data processing speed of discrete cosine conversion integration module 1 of the present invention improves greatly.
More than describing only is a specific embodiment of the present invention, in fact discrete cosine conversion integration module 1 of the present invention can also be made the change of some equivalences according to foregoing description, for example, shown in 3 kinds in figure, selector 150,151,152 connected mode is not limited to the connected mode described in Fig. 1, as long as can reach control effect of the present invention, selector 150,151,152 connected modes that can adopt other to be fit to fully, promptly, can realize image data storage apparatus 130 with a suitable selector unit 5, quantification/re-quantization device 140, discrete cosine transform/inverter 160 is selected data, be configured and control by control device 110, and then reach goal of the invention of the present invention.
All should be included within the scope that claim of the present invention protects such as the minor alteration of above-mentioned situation and equivalent transformation.
Claims (8)
1, discrete cosine conversion integration module comprises:
Input/output device is used for exchange and quantizes coefficient table, view data and control information;
Quantification/re-quantization coefficient table storage device, storage quantizes coefficient table;
Discrete cosine transform/inverter is used for view data is carried out discrete cosine transform/inverse transformation;
Quantification/re-quantization device is used to read the quantization parameter table and view data is quantized/the re-quantization computing;
It is characterized in that:
Described discrete cosine conversion integration module further comprises
Image data storage apparatus;
Selector unit can be selected data for image data storage apparatus, quantification/re-quantization device, discrete cosine transform/inverter; And
Control device is used to control selector unit, to image data storage apparatus, quantification/re-quantization device, discrete cosine transform/inverter is configured and control.
2, according to the described discrete cosine conversion integration module of claim 1, it is characterized in that:
Described image data storage apparatus has two can be by the storage area of concurrent access.
3, discrete cosine conversion integration module according to claim 1 and 2 is characterized in that:
Described selector unit comprises three independently selectors (150,151,152), and each selector includes two inputs and an output.
4, discrete cosine conversion integration module according to claim 3 is characterized in that:
The output of described selector (150) connects image data storage apparatus, and two input connects quantification/re-quantization device and discrete cosine transform/inverter respectively;
The output of described selector (151) connects quantification/re-quantization device, and two input connects image data storage apparatus and discrete cosine transform/inverter respectively;
The output of described selector (152) connects discrete cosine transform/inverter, and two input connects image data storage apparatus and quantification/re-quantization device respectively.
5, according to the computing combination method of the described discrete cosine conversion integration module of claim 1, it may further comprise the steps:
The first step, control device starts computing, reads the data that need conversion from image data storage apparatus;
In second step, judge whether to carry out inverse discrete cosine transform:
If do not carry out inverse discrete cosine transform, then enter data into and carry out discrete cosine transform in discrete cosine transform/inverter, after conversion finishes, judge whether to carry out quantization operations, if do not quantize, then directly result data is deposited in the image data storage apparatus, carry out quantization operations if desired, then quantification/re-quantization device carries out quantization operation according to the control information of control device, and then result data is deposited in the image data storage apparatus;
Carry out inverse discrete cosine transform if desired, then further judge whether to carry out the re-quantization computing, if do not need to carry out the re-quantization computing, then carry out inverse discrete cosine transform by discrete cosine transform/inverter, and the inverse transformation result data deposited in the image data storage apparatus, carry out the re-quantization computing if desired, quantification/re-quantization device carries out the re-quantization operation according to the control information of control device, carry out inverse discrete cosine transform by discrete cosine transform/inverter again, deposit result data in image data storage apparatus then.
6, the computing combination method of discrete cosine conversion integration module according to claim 5, it further may further comprise the steps: after executing for second step, judge that whether total transform block number is finished, if do not finish, then re-executes the first step; If total transform block number is finished, then send interrupt signal or set end mark position, and finish computing by control device.
7, according to the computing combination method of claim 5 or 6 described discrete cosine conversion integration modules, it is characterized in that:
Described image data storage apparatus is divided into two storage areas, under the control of control device, when input/output device is visited one of them storage area, another storage area can be by discrete cosine transform/inverter or quantification/re-quantization device visit.
8, the computing combination method of discrete cosine conversion integration module according to claim 7 is characterized in that:
The execution in step of described discrete cosine transform/inverse transformation or quantification/re-quantization by three independently selector finish.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200610035021 CN101018327B (en) | 2006-04-11 | 2006-04-11 | Discrete cosine conversion integration module and its computing combination method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200610035021 CN101018327B (en) | 2006-04-11 | 2006-04-11 | Discrete cosine conversion integration module and its computing combination method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101018327A true CN101018327A (en) | 2007-08-15 |
| CN101018327B CN101018327B (en) | 2012-10-31 |
Family
ID=38727052
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200610035021 Expired - Fee Related CN101018327B (en) | 2006-04-11 | 2006-04-11 | Discrete cosine conversion integration module and its computing combination method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101018327B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101540172B (en) * | 2009-04-15 | 2012-02-22 | 炬力集成电路设计有限公司 | Encoding and decoding system of audio data and method thereof |
| CN105491389A (en) * | 2009-06-05 | 2016-04-13 | 高通股份有限公司 | 4X4 transform for media coding |
| CN117111637A (en) * | 2023-09-26 | 2023-11-24 | 江苏星湖科技有限公司 | Method and device for controlling patrol of convergence and fusion terminal of power high-altitude sensing equipment |
| CN117111637B (en) * | 2023-09-26 | 2024-05-10 | 江苏星湖科技有限公司 | Method and device for controlling patrol of convergence and fusion terminal of power high-altitude sensing equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100323441B1 (en) * | 1997-08-20 | 2002-06-20 | 윤종용 | Mpeg2 motion picture coding/decoding system |
| JP2005303420A (en) * | 2004-04-07 | 2005-10-27 | Alps Electric Co Ltd | Television signal transmitter |
| CN2689607Y (en) * | 2004-04-12 | 2005-03-30 | 南京师范大学 | Continuous phase pai fourth differential coded orthogonal phase-shifting key controlling modem device |
-
2006
- 2006-04-11 CN CN 200610035021 patent/CN101018327B/en not_active Expired - Fee Related
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101540172B (en) * | 2009-04-15 | 2012-02-22 | 炬力集成电路设计有限公司 | Encoding and decoding system of audio data and method thereof |
| CN105491389A (en) * | 2009-06-05 | 2016-04-13 | 高通股份有限公司 | 4X4 transform for media coding |
| CN105744280A (en) * | 2009-06-05 | 2016-07-06 | 高通股份有限公司 | 4x4 transforms for media coding |
| CN105491389B (en) * | 2009-06-05 | 2018-12-04 | 高通股份有限公司 | The equipment that 4X4 for implementing for media coding is converted |
| CN117111637A (en) * | 2023-09-26 | 2023-11-24 | 江苏星湖科技有限公司 | Method and device for controlling patrol of convergence and fusion terminal of power high-altitude sensing equipment |
| CN117111637B (en) * | 2023-09-26 | 2024-05-10 | 江苏星湖科技有限公司 | Method and device for controlling patrol of convergence and fusion terminal of power high-altitude sensing equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101018327B (en) | 2012-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7236525B2 (en) | Reconfigurable computing based multi-standard video codec | |
| CN102369552B (en) | Memory subsystem | |
| EP0637173A2 (en) | Digital signal processor | |
| CN100369024C (en) | Direct memory access control device and image processing system and transmission method | |
| JP2010527194A (en) | Dynamic motion vector analysis method | |
| CN102017625A (en) | Decoding device | |
| CN101146222B (en) | Motion estimation core of video system | |
| Kammoun et al. | Design exploration of efficient implementation on SoC heterogeneous platform: HEVC intra prediction application | |
| CN101729893A (en) | MPEG multi-format compatible decoding method based on software and hardware coprocessing and device thereof | |
| CN102572430A (en) | Method for implementing H.264 deblocking filter algorithm based on reconfigurable technique | |
| CN101018327B (en) | Discrete cosine conversion integration module and its computing combination method | |
| CN1703094A (en) | Image interpolation apparatus and methods that apply quarter pel interpolation to selected half pel interpolation results | |
| KR100564010B1 (en) | Image processor | |
| CN110324632A (en) | A kind of data processing and verification method based on OpenMP multi-core parallel concurrent mechanism | |
| CN101383968A (en) | Video decoder, video decoding method and mobile multimedia terminal chip | |
| CN103975583A (en) | Capturing multiple video channels for video analytics and encoding | |
| CN103765888A (en) | Analytics assisted encoding | |
| CN100486333C (en) | Interpolation arithmetic device and method | |
| De Souza et al. | OpenCL parallelization of the HEVC de-quantization and inverse transform for heterogeneous platforms | |
| Dyer et al. | Design and analysis of system on a chip encoder for JPEG2000 | |
| Tumeo et al. | Prototyping pipelined applications on a heterogeneous fpga multiprocessor virtual platform | |
| US20050080784A1 (en) | Data processing system | |
| Kim et al. | Hardware‐Software Implementation of MPEG‐4 Video Codec | |
| Hilgenstock et al. | A video signal processor for MIMD multiprocessing | |
| CN115145842A (en) | Data cache processor and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20170612 Address after: 519085 C District, 1# workshop, No. 1, science and technology No. four road, hi tech Zone, Zhuhai, Guangdong, China Patentee after: ACTIONS (ZHUHAI) TECHNOLOGY CO., LTD. Address before: 519085 hi tech Zone, Guangdong, Zhuhai science and Technology Innovation Coast Road, No. four, No. 1 Patentee before: Juli Integrated Circuit Design Co., Ltd. |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20121031 Termination date: 20180411 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |