CN1547856A - Scalable programmable motion image system - Google Patents
Scalable programmable motion image system Download PDFInfo
- Publication number
- CN1547856A CN1547856A CNA028077148A CN02807714A CN1547856A CN 1547856 A CN1547856 A CN 1547856A CN A028077148 A CNA028077148 A CN A028077148A CN 02807714 A CN02807714 A CN 02807714A CN 1547856 A CN1547856 A CN 1547856A
- Authority
- CN
- China
- Prior art keywords
- moving image
- compression
- module
- motion image
- digital motion
- 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
- 230000033001 locomotion Effects 0.000 title claims abstract description 108
- 238000007906 compression Methods 0.000 claims abstract description 58
- 230000006835 compression Effects 0.000 claims abstract description 58
- 230000015654 memory Effects 0.000 claims abstract description 25
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 20
- 230000005540 biological transmission Effects 0.000 claims abstract description 14
- 230000011218 segmentation Effects 0.000 claims description 9
- 238000013139 quantization Methods 0.000 abstract description 4
- 230000002123 temporal effect Effects 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 22
- 238000010586 diagram Methods 0.000 description 18
- 239000000872 buffer Substances 0.000 description 15
- 238000004040 coloring Methods 0.000 description 12
- 230000008859 change Effects 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 9
- 238000006243 chemical reaction Methods 0.000 description 8
- 230000008878 coupling Effects 0.000 description 8
- 238000010168 coupling process Methods 0.000 description 8
- 238000005859 coupling reaction Methods 0.000 description 8
- 238000000034 method Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 230000009466 transformation Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000001914 filtration Methods 0.000 description 5
- 238000011002 quantification Methods 0.000 description 5
- 230000001360 synchronised effect Effects 0.000 description 5
- 230000003139 buffering effect Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000006837 decompression Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 244000287680 Garcinia dulcis Species 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 235000021384 green leafy vegetables Nutrition 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000007115 recruitment Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/13—Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
- H04N19/436—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation using parallelised computational arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/103—Selection of coding mode or of prediction mode
- H04N19/112—Selection of coding mode or of prediction mode according to a given display mode, e.g. for interlaced or progressive display mode
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/154—Measured or subjectively estimated visual quality after decoding, e.g. measurement of distortion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/162—User input
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/1883—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit relating to sub-band structure, e.g. hierarchical level, directional tree, e.g. low-high [LH], high-low [HL], high-high [HH]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/46—Embedding additional information in the video signal during the compression process
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/62—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding by frequency transforming in three dimensions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
- H04N19/63—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding using sub-band based transform, e.g. wavelets
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
A scalable motion image compression system for a digital motion image signal having an associated transmission rate. The scalable motion image compression system includes a decomposition module for receiving the digital motion image signal, decomposing the digital motion image signal into component parts and sending the components. The decomposition module may further perform color rotation, spatial decomposition and temporal decomposition. The system further includes a compression module for receiving each of the component parts from the decomposition module, compressing the component part, and sending the compressed component part to a memory location. The compression module may perform sub-band wavelet compression and may further include functionality for quantization and entropy encoding.
Description
Technical field
The present invention relates to digital motion picture, particularly a kind of architecture that is used for digital motion image system being carried out convergent-divergent according to various digital motion picture forms.
Background technology
In half a century in the past, single format professional and consumer video record device have developed into the sophisticated systems that has by the desired specific function of film makers and videographer.Along with the appearance of high definition digital imaging, many motion image formats are not having sharply increase under the standardized situation.Along with the exploitation of digital imagery, designed the digital data compression technology, with the realization higher resolution image, thereby can be stored in more information in the same memory space as unpressed lower resolution image.In order to realize the storage of higher resolution image, record with manufacturer of memory devices and compress technique is added in its system.In general, current compress technique is to use discrete cosine transform (DCT) that each image in the video sequence is carried out space encoding.The intrinsic property of this processing is the following fact, that is: space encoding is based on piece.This block-based system is not easy to realize scalability due to the fact that, and this fact is that along with the increase of image resolution ratio, amount of compressed data also increases pro rata.This piece transformation system can not be seen correlation on block boundary or with the frequency that is lower than block size.Because the low frequency that exemplary power distributes biasing, along with the increase of image size, increasing information will be positioned at the below horizontal of piece conversion.Therefore, the piece transform method that is used for the spatial image compression will tend to generate the data volume that is in proportion with image according to given quality.And along with the increase of resolution, the tiling effect (tiling effects) that is caused by block-based coding becomes more remarkable, thereby has the great amount of images loss, comprises image abnormality and discontinuity.Because these restrictions, manufacturer has designed its compressibility at narrow resolution.For each resolution of already expecting by film, forced these manufacturers to overcome these defectives again, and the application program of exploitation resolution specific come the space encoding problem is remedied.As a result, the also untapped image display system that goes out to carry out at moving picture stream convergent-divergent with different throughputs.
Summary of the invention
The present invention has disclosed a kind of scalable moving image compressibility that is used to have the digital motion image signal of associated transmission rate.This scalable moving image compressibility comprises: decomposing module, be used to receive digital motion image signal, and this digital motion image signal is resolved into a plurality of components, and send these components.This decomposing module can also be carried out look rotation, spatial decomposition and time and be decomposed.This system also comprises compression module, is used to receive each component from decomposing module, this component is compressed, and this compressed component is sent to memory cell.This compression module can carry out subband (sub-band) small echo (wavelet) compression, and can also comprise quantification and entropy coding function.
Each decomposing module all can comprise it can being one or more resolving cells of asic chip.Equally, each compression module all can comprise it can being one or more compression units of CODEC asic chip.
This system can compress the digital motion image stream of being imported according to transmission rate in real time.This system can also comprise programmable module, is used for the digital motion image signal after Route Selection between decomposing module and the compression module is decomposed.This programmable module can be the field programmable gate array as router.In this embodiment, decomposing module has one or more resolving cells, and compression module has one or more compression units.
In another embodiment, field programmable gate array is a Reprogrammable.In another embodiment, resolving cell adopts and to be set up in parallel, and each unit all receives the part of the digital motion image signal stream of being imported, and makes the total throughout of resolving cell greater than the transmission rate of digital motion image stream.In certain embodiments, decomposing module is constituted as by look, frame or field digital motion image stream is decomposed.Decomposing module can also be carried out look decorrelation (colordecorrelation).Decomposing module and compression module all are Reprogrammables, and have the memory that is used for for the reception coefficient value of coding and filtering use.It will be understood by those skilled in the art that this system can be used for the digital motion image stream of compression is decompressed equally.Each module all can receive one group of new coefficient, thereby may be implemented as inverse filter.
Description of drawings
Detailed description by doing with reference to following concrete accompanying drawing will be easier to understand These characteristics of the present invention, in the accompanying drawings:
Fig. 1 is the block diagram that the one exemplary embodiment of scalable video of the present invention system is shown;
Fig. 2 is the block diagram that is depicted as a plurality of digital motion image system chip that generate scalable digital motion image system and be coupled;
Fig. 2 A illustrates the flow chart of digital motion image stream by digital motion image system;
Fig. 2 B shows a group of module;
Fig. 3 illustrates the block diagram that can be arranged on the various modules on the digital motion image chip;
Fig. 4 is the block diagram that is illustrated in the synchronous communication scheme between DMR and the CODEC;
Fig. 5 shows and is used for synchronizing signal is offered each DMR and CODEC in the single chip, and when adopting array format to connect, can offer synchronizing signal by the bus interface module (not shown) block diagram of the global control module of all chips in this array;
Fig. 6 is the block diagram that is illustrated in an example of the preceding digital motion image system chip of configuration;
Fig. 7 A and Fig. 7 B are the block diagrams of functional part that the digital motion image system chip of the Fig. 6 after the configuration is shown;
Fig. 8 illustrates to be arranged on the interior element of CODEC and the block diagram of bus;
Fig. 9 is the block diagram that the heterogeneous processing example in space is shown; And
Figure 10 illustrates the block diagram that the spatial subbands of using DMR and CODEC is cut apart example.
Embodiment
Definition: unless context has requirement in addition, the following term that uses in this specification and claims should have the regulation implication:
Pixel is a pictorial element, and generally is the minimum controllable look element on the display unit.Pixel is relevant with look information in the concrete color space.For example, digital picture can have 640 * 480 pixel resolution in RGB (redness, green, the blueness) color space.This image has in 480 row and has 640 pixels, and wherein, each pixel all has relevant red value, green value and blue valve.Moving picture stream can be made of digital data stream, and this digital data stream can be divided into a plurality of or frame of expression moving image, and wherein, a frame is meant the complete image that will be presented at the numerical data on the display unit in a time durations.Moving image frame can be broken down into a plurality of.A field generally is designated as odd field or even field, this means during preset time in all odd lines or all even lines of display image.Show in during different time that even field and odd field are called as staggered in this area.It will be understood by those skilled in the art that a frame or a pair of expression complete image.The term of Shi Yonging " image " should refer to field and frame in this article.And, in this article the term of Shi Yonging " Digital Signal Processing " Ying Zhiwei data stream is changed and/or segmentation and according to organizational form to digital data stream handle.
Fig. 1 is the block diagram that the one exemplary embodiment of scalable video of the present invention system 10 is shown.This system comprises digital video system chip 15, is used for digital motion image stream is received in the input 16.This digital motion image system chip 15 is implemented preferably as the integrated circuit (ASIC) of application-specific.Be used for processor that digital motion image system chip is controlled 17 and provide instruction to digital motion image system chip, these instructions can comprise various instructions, for example, Route Selection, compression stage is set, and coding, it comprises room and time coding, look decorrelation, color space transformation, alternation sum is encrypted.15 pairs of digital motion image stream 16 of digital motion image system chip are compressed, thus the generation digital data stream 18 of near real time, and this information send to memory be used for after the retrieval.Processor can be filed a request to digital motion image system chip, and this digital motion image system chip will flow to line retrieval to digital data and make and handle oppositely, so that digital motion image stream is output 16.Digital motion image stream is sent to numeric display unit 20 from this output.
Fig. 2 is the block diagram that is depicted as a plurality of digital motion image system chip 15 that generate a kind of scalable digital motion image system and be coupled, this scalable motion image system can hold various digital motion image stream, and each digital motion image stream all has dependent resolution and relevant throughput.For example, digital motion image stream can have the resolution of every moving image 1600 * 1200 pixels, and each pixel is represented by 24 information (8 redness, 8 greens, 8 bluenesss), and this digital motion image stream can have the speed of per second 30 frames.This moving picture stream will need a kind of device of throughput of the peak rate that can have 1.38 gigabit/sec.This system can hold various resolution by various configurations, comprising: for example, and 640 * 480,1280 * 768 and 4080 * 2040.
Fig. 2 A shows the method for carrying out this processing.At first, digital motion image stream is received in this system.According to throughput, this stream is separated such as frame or the such definable point of line point in image, and is assigned to a chip in a plurality of chips, makes chip that buffer is provided, with the throughput (step 201A) that adapts to digital motion image stream.Then, these chips for example decompose image stream by colouring component or by the field separately.Then, these chips will carry out decorrelation (step 202A) to digital picture stream according to decomposition.For example, can carry out decorrelation to isolate brightness, also can carry out conversion sub-band coding (transform sub-band code) each image (field/frame) in this stream to colouring component.Then, this system encodes to this stream by quantification and entropy coding, so that further the data volume of his-and-hers watches registration word moving image is compressed (step 203A).Below will be described further these steps.
If the parts on the digital motion image system chip can not provide this peak throughput separately, then can be with electric parallel connection of these chips and/or series connection, so that by at first digital motion image stream being cushioned, then digital motion image stream is resolved into picture content, and between other motion image system chip, redistribute these components necessary throughput is provided.This decomposition can use register input buffer to realize.For example, if necessary throughput is the twice of digital motion image chip capacity, two registers that then setting had the word length of moving picture stream, so that will be placed into data in the register according to suitable frequency, but will read this data from register according to half or two word lengths of every circulation of this frequency.And, a plurality of digital motion image system chip are linked to form a kind of like this buffer.Suppose that switch can operate according to the speed of digital motion image stream, then each digital motion image system chip can receive the part of this stream, and can the part of this stream be cushioned.For example, suppose that digital motion image stream is made of 4000 * 4000 pixel monochrome images with per second 30 frames.Required throughput is 4.8 hundred million components of per second.If digital motion image system chip only has the maximum throughput of 0.6 hundred million components of per second, then this system can constitute the chip chamber of switch in 8 chips of operating according to 4.8 hundred million components of per second switched in turn.Then, digital video system chip is used as buffer with each.As a result, digital motion image stream then can obtain handling in chip.For example, the frame ordering can be changed, and perhaps this system can set up or eliminate pixel, field or the frame of data.
After buffering, digital motion image stream is decomposed.For example, digital motion image system chip can provide look to decompose, so that each moving image is separated into its colouring component separately, and for example RGB or YUV colouring component.During decomposing, also can carry out decorrelation to signal.Rely on rotation of coordinate, can check colors and carry out decorrelation, so that monochrome information and look information are isolated.Its allochromatic colour is decomposed and decorrelation also is feasible.For example, can carry out decorrelation and decomposition to earth resource (EarthResources) image of 36 components, wherein, each component is all represented frequency band, thereby makes spatial information relevant with look information.Generally, these components are shared these two kinds of public monochrome informations, and also have the effective correlation with the advancing colour component.In the case, small echo (wavelet) conversion can be used for these components are carried out decorrelation.
In many digital image stream formats, look information is mixed mutually with space and frequency information, and for example, the look of colouring component only being taken a sample at each pixel cell covers (mask) imager.In this case, the look decorrelation needs space and frequency decorrelation.For example, the camera of supposing 4000 * 2000 pixels uses 3 looks to cover (green, redness is in 2 * 2 repeated grid for blueness, green), and operates according to the frame rate that can reach 72Hz.Then, this camera will provide per second can reach 5.76 hundred million simple component pixels.Suppose that System on Chip/SoC can be imported 600,000,000 components and but per second is handled 300,000,000 components, then two System on Chip/SoCs can be used as heterogeneous frame buffer, and can make four phase acoustic convolvers pass through data according to per second 300,000,000 components.Each of acoustic convolver is all corresponding with during look covers one mutually, and generate four isolated components as output, that is: two dimension half is with (half band) low-frequency brightness component, and two dimension partly is with the high frequency diagonal luminance component, and two dimension partly is with Cb color difference components and two dimension partly to be with the Cr color difference components.The inromation bandwidth of this processing is retained, and wherein, generates four independent equiband components, and the color space is carried out decorrelation.(bandlimiting) is relevant with subband solutions includes in the single heterogeneous convolution interpolation, color space decorrelation, limit band for just described two-dimensional convolution device.It will be understood by those skilled in the art that further decomposition is feasible.Because the modularization of digital motion image system, thereby these various types of decorrelations are feasible with decomposing.As further described below, each element of chip all obtains external control and configurable.For example, have individual component in chip, be used to carry out look decomposition, space encoding and time encoding, wherein, each conversion is designed to the many tap filters according to the coefficient value definition.Ppu can be imported different coefficient values at concrete element according to application program.And ppu can be selected will be for the related elements of handling usefulness.For example, digital motion image system chip can only be used for buffering and look decomposes, and only is used for space encoding, perhaps is used for the room and time coding.This modularization in the chip is partly realized by the bus with each element coupling.
By frame being separated into a plurality of, can also decompose moving image.For example constitute, can also or decompose frame according to the frequency of image, so as to image low, in and high frequency components make up.It should be appreciated by those skilled in the art that other frequency segmentations also are feasible.Should also be noted that the decomposition of being mentioned is non-spatial decomposition, thereby when decompressing, eliminate the discontinuity in the reconstructing digital moving picture stream, these discontinuities are general in block-based compress technique.As described, as the result of the decorrelation of digital motion image stream, owing to parallel processing makes total throughout can bring up to N doubly.For example, in following example, N will be 27: 1, and wherein, image is divided into a plurality of (gains in 2: 1), is divided into a plurality of colouring components (3: 1) gain then, is divided into a plurality of frequency components (3: 1) gain then.Therefore, total recruitment of throughput is 27: 1, thereby can realize taking place the final processing of actual compression and coding according to 1/27 speed of the speed of institute's input motion image stream.Therefore, can carry out convergent-divergent to the throughput relevant with image resolution ratio.In this embodiment, because motion image chips has the I/O capacity of simple staggered 1.3G component/second of decomposing, thereby a pair of motion image chips can be connected the output port of first motion image chips, can in second pair of motion image chips, carry out color component decomposition then, wherein, the look decomposition is no more than 650 megabit per seconds, therefore, can keep total throughout.In addition, can realize further decomposing on the basis frame by frame, general, this be called as heterogeneous processing (poly-phasing) in the art.
Digital motion image stream self can enter motion image chips by a plurality of channels.For example, can carry out segmentation to the Quad-HD signal by 8 channels.In this constitutes, can use 8 independent digital motion image chip to be used for digital motion image stream is compressed, each channel is equipped with one.
Each moving image all has: I/O (I/O) port or lead-in wire (pin) are used for providing data at chip chamber; And data communication port, be used in the chip chamber transmission that gives information.Should be appreciated that processor is controlled chip array by providing for each chip in the chip array and the relevant instruction of Digital Signal Processing task that will carry out digital motion image data.And, should be appreciated that, the memory input/output end port is set on each chip, be used for communicating with storage arbiter (arbiter) and memory cell.
In one embodiment, each digital motion image system chip all comprises input/output end port and a plurality of module, and these a plurality of modules comprise decomposing module 25, field programmable gate array (FPGA) 30 and compression module 35.Fig. 2 B shows a grouping of module.In a practical embodiments, a plurality of this groupings will be included on the single chip.Therefore, FPGA allows chip is programmed, so that the coupling between decomposing module and the compression module is configured.
For example, by each frame of moving picture stream is divided into its colouring component separately, can in decomposing module, decompose the moving data streaming of being imported.FPGA can be the FPGA of dynamic Reprogrammable, it will be programmed to be used to receive three moving-picture information streams, and (in this embodiment, one is used for redness, and one is used for green, one is used for blueness) multiplexer/router, and this information is sent to compression module.Although field programmable gate array is described, yet can use other signal/data distributors.Distributor can use mark (token) to send distributing signal on reciprocity basis, perhaps distributor can obtain center control, and independent distributing signal, perhaps distributor can offer each module to whole moving image input signal, thereby the part of the module that should not obtain handling is covered.Then, constitute and each compression unit all can will compress this stream the compression module that inlet flow compresses by a plurality of compression units, and this packed data is preferably outputed to memory.The compression module of this preferred embodiment adopts wavelet compression, and this wavelet compression is used sub-band coding to this stream in room and time.This compression module further is set so that according to the control signal that sends to compression module from processor, for compression in various degree give security the level signal quality.Therefore, this compression module generates the signal of compression, in case decompress, this compressed signal just keeps setting resolution in all frequency ranges at the image sequence in the digital motion image stream.
If the component processing speed m of System on Chip/SoC is less than n, wherein, n represents isolated component speed, then uses Roof[m/n] System on Chip/SoC.Each System on Chip/SoC all receives each Roof[n/m] pixel or Roof[n/m] frame.This selection is to determine according to the fluency (ease) of I/O buffering.For Roof[n/m] not that the pixel of multiple of line length of the video image in handling is heterogeneous, the circuit liner is used to keep vertical correlation.Relying under the multiplexing heterogeneous situation of component, keep vertical correlation, and can make sub-band transforms be applicable to image column in the each several part independently, to produce two or more quadratures segmentations of vertical component.Relying under the multiplexing heterogeneous situation of frame, kept vertical and horizontal correlation, thereby can make the two-dimensional sub-band conversion be applicable to frame, to generate two or more quadrature segmentations of two-dimensional signal.System on Chip/SoC is designed to be supported in the identical peak rate of input port and output port.This Roof[n/m] non-heterogeneous, the subband that handle to adopt (transposed) heterogeneous mode of transposition to export input signal represents wherein, more components are arranged now, and each isolated component adopts the speed that reduces.
Fig. 3 shows the various modules that can be arranged on the digital motion image chip 15, comprising: decomposing module 300, this decomposing module 300 can comprise one or more resolving cells 305.Colorimetric compensation, color space rotation, look decomposition, room and time conversion, format conversion can be realized in these unit, and other moving image Digital Signal Processing functions.And this resolving cell 305 can be called as the heavy formatter (Digital MasteringReformatter) (" DMR ") of digital master control.This DMR 305 also is provided with " intelligence " I/O port, should " intelligence " I/O port generally use space, time and look decorrelation that single tap or two-tap filter realize simplifying, look rotation, by interpolation and select (decimation) the position convergent-divergent, 3:2 is drop-down and the line multiplication.This intelligence I/O port is bidirectional port preferably, and is provided with the application specific processor that is used to receive command sequence.This input port and output port all are set to separate operation, and like this, for example, input port can carry out the time decorrelation of colouring component, and output port can carry out the interweaving of line of each image and moves sliding (shuffle).The META data that the instruction of this I/O port can be used as in the digital motion image stream are transmitted, also can be sent to the I/O porthandler by system processor, wherein, system processor is the processor that does not belong to a digital motion image chip part, and instruction offered is used for chip that chip functions is controlled.The I/O port also can be used as standard I/O port, and can be sent to the internal applications dedicated digital signal processor to numerical data, and this processor carries out high-order filtering.I/O processor and system clock are synchronous, and like this, in case the synchronization of time intenals of appointment finishes, the I/O port just will under normal circumstances preferably be sent to next module to the deal with data of entire frame, and receives the data of another frame of expression.If synchronization of time intenals finishes, and the data in the module are not handled fully, and then output port will still be removed half data of handling, and input port will receive next and organize data.For example, if the throughput of digital motion image stream surpasses the throughput of single DMR305 or compression module, then will walk abreast and use DMR 305 and used as buffer.In this constituted, when switch/signal splitter (partitioner) was input to numerical data in each DMR, this DMR can further decompose and/or decorrelation.
In certain embodiments, CODEC comprises the ability of interlocking and handling and encrypting.CODEC also has " intelligence " I/O port, should " intelligence " I/O port can use the decorrelation of simplifying such as single tap and the such simple filter of two-tap filter, and the employing mode identical with the above-mentioned intelligent I/O port of DMR operated.This DMR and CODEC are provided with the input and output buffer, these buffers are provided for receiving from the digital motion image stream of another DMR or CODEC or the memory cell of data, and be used to be stored in handle after but the unit of data before being transferred to DMR or CODEC.In this preferred embodiment, for DMR and CODEC, input and output port has same band, but for the support module scheme, also needn't have same band.For example, preferably, the I/O speed height of the I/O speed ratio CODEC of this DMR is to support heterogeneous buffering.Because each CODEC has same band at input port and output port, thereby can connect CODEC by the common bus lead-in wire easily, and can use public clock that CODEC is controlled.
And as described in the 09/498th, No. 924 U.S. Patent application book, CODEC can constitute in the quality priority mode and operate, and this U.S. Patent application book is quoted in this article comprehensively, for your guidance.In quality priority, each frequency band that has used the subband wavelet transformation to carry out the frame of video of decorrelation can have quantization level, this quantization level sampling theory curve in the information faces of videoing.This curve has resolution axis and frequency axis, and for each octave in Nyquist (Nyquist) frequency downstream, needs additional 1.0 to represent two dimensional image.The resolution of the video flowing of representing when therefore, in all frequency ranges, remaining on nyquist frequency.According to sampling theory,, need additional 1/2 of every dimension resolution for each octave in downstream.Therefore, when lower frequency, need the more information of multidigit, with the expression resolution identical with Nyquist.Therefore, the peak rate during quantification can be near the data rate in the sample domain, and therefore, the input port of CODEC and output port should have approximately uniform throughput.
Because high-definition picture can be broken down into and can can't influence a plurality of than junior unit of picture quality with the throughput compatibility of CODEC, thereby can carry out the additional character signal processing to image, for example, homomorphism (homomorphic) filtering and granularity reduce.Can change quantification according to for example people's perception, sensor resolution and equipment energy characteristic.
Therefore, this system can adopt modular multiplexed form (multi-plexed) to constitute, and it has fixing throughput to adapt to the module of different images size.This system can realize this point, and can not lose owing to horizontal effect and blocking artefacts (artifacts), and this is because compression is to be transformed to the basis with the full images that the part is supported.This system also can carry out pyramid (pyramid) conversion, so that further sub-band coding is carried out in bottom and bottom frequency component.
The various structures that it will be understood by those skilled in the art that this CODEC and DMR can be arranged on the single motion image chips.For example, this chip can be specially constituting by multiplexing CODEC, multiplexing DMR or DMR and CODEC.And digital motion image chip can be single CODEC or single DMR.The processor that is used for digital motion image system chip is controlled can provide control command, makes chip use multiple CODEC to carry out N component colour coding, variable frame-rate coding (for example, per second 30 frames or per second 70 frames), and the high-resolution coding.
Fig. 3 also shows the connection between DMR 305 and compression module 350, makes DMR to be used for parallel processing to the each side that decomposed information sends among a plurality of CODEC 355.Should be appreciated that the not shown FPGA/ signal distributor of this figure.In case FPGA is programmed, this FPGA is signalization path between suitable decomposing module and compression module just, thereby as signal distributor.
Fig. 4 is the block diagram that is illustrated in the synchronous communication scheme between DMR 400 and the CODEC 410.Message transmission between these two unit is provided by signaling channel.This DMR 400 sends signal to CODEC 410, represents that it prepare to use READY (preparations) to instruct 420 information write CODEC.Then, DMR waits for that CODEC uses WRITE (writing) instruction 430 to make and replys.When receiving that WRITE instructs 430 the time, this DMR is sent to CODEC to next data cell, promptly is sent in the CODEC input buffer from the DMR output buffer.What this CODEC also can make NOT READY (prepare) replys 440, and then, DMR will wait for that CODEC use READY signal 420 is made and reply, thereby data be remained in the output buffer of DMR.In this preferred embodiment, when the input buffer of CODEC is in 32 full words, CODEC will send NOT READY and reply 440.When DMR received NOT READY 440, DMR stopped the current data unit is handled.This handshaking at intermodule is standardized, makes each decomposing module and each compression module all can understand signal.
Fig. 5 shows and is used for synchronizing signal 501 is offered each DMR 510 and CODEC 520 in the single chip, and when adopting array format to connect, can offer synchronizing signal by the bus interface module (not shown) block diagram of the global control module 500 of all chips in the array.In this preferred embodiment, synchronizing signal takes place according to a frame rate of moving image, and then, synchronizing signal can take place according to the speed of a group image information.For example, if according to the speed of per second 24 frames the digital motion image stream of being imported is taken, then synchronizing signal took place per 1/24 second.Therefore,, transmit information, make DMR adopt the decorrelation form whole frame of digital motion picture to be sent to the compression module of CODEC at intermodule in each synchronizing signal.Similarly, new digital moving picture frame is sent to DMR.This global synchronizing signal has precedence over and is included in the READ (reading) that transmits between DMR and the CODEC and WRITE (writing) instruction at interior every other signal.Therefore, READ and WRITE instruction belongs to during the interframe.Synchronizing signal forces and transmits a group image information (frame in the preferred embodiment), so that frame remains on synchronous regime.Grow during between synchronizing signal if CODEC handles institute's time spent to a group image information, then this group image information goes out of use, and the data of all section processes in DMR or the CODEC are eliminated.Global synchronizing signal is transmitted along global control bus, and this global control bus is shared by all DMR on the chip and CODEC or adopted array format to constitute.This overall situation control also comprises global direction signal.Global direction signal represents to the I/O of DMR and CODEC port whether this port should send or receive data.By regularly scheme of synchronizing signal is provided, can keep the throughput of system, therefore, scalable system acting unanimity, and thereby can be from such as the such soft error of the instantaneous noise of arbitrary component inside or such as recovering the such external error of misdata.
Fig. 6 is the block diagram that an example of digital motion image system chip 600 is shown.This chip is provided with a DMR 610, follows by FPGA 620, is a pair of DMR 630A-B subsequently, and this is coupled with the 2nd FPGA 640A-B separately to DMR 630A-B.And the coupling of the each side among this FPGA and four the CODEC 650A-H.As mentioned above, this FPGA can programme according to the expectation throughput.For example, in Fig. 7 A, a FPGA 620 is configured to make it to be coupling between a DMR 610 and the 2nd DMR 630A.The 2nd DMR 630A and FPGA640A coupling, and FPGA 640A and three CODEC 650A, 650B, 650C are coupled.This structure can be used to input digital image stream is divided into a plurality of frames among the DMR, and the colouring component to each frame among the 2nd DMR carries out decorrelation then.CODEC in the present embodiment compresses the data of a colouring component of each moving image frame.Fig. 7 B is the alternative structure of the digital motion image system chip of Fig. 6.In the structure of Fig. 7 B, a FPGA620 is configured to make its each coupling in output and two DMR 630A, 630B.Then, each DMR 630A, B send to single CODEC 650A, 650E to data.This structure can at first be used to make moving image frame to interweave, and makes the 2nd DMR receive odd field or even field.Then, the 2nd DMR can carry out that look is proofreaied and correct or color space transformation to the digital moving picture frame after interweaving, and then, these data is sent to single CODEC, this CODEC to proofread and correct through look interweave after digital motion picture compress and encode.
Fig. 8 illustrates the element that is arranged in the CODEC 800 and the block diagram of bus.The element of DMR can be equal to the element of CODEC.This DMR preferably has more based on multiple data rates throughput, is used for receiving higher component/second digital motion image stream, and also has more internal memories, is used for the reception data of digital motion image stream are cushioned.This DMR can constitute and only carry out the color space and spatial decomposition, makes DMR have data I/O port and image I/O port, and is coupled with memory, and wherein, these I/O ports comprise the programmable filter that is used to decompose.
This CODEC 800 and global control bus 810 couplings, this global control bus 810 communicates with each element under control.These elements comprise: data I/O port 820, encryption element 830, encoder 840, spatial transform element 850, time change element 860, interleaving treatment element 870 and image I/O port 880.All elements all are coupled by the public multiplexer (mux) 890 with memory 895 couplings.In this preferred embodiment, memory is double data rate (DDR) memory.Each element all can be independent of every other element to be operated.Global control module is to sending command signal to the element that data flow to line number word signal processing.For example, global control module can only communicate with spatial transform element, makes that stream only carries out spatial alternation to digital data.In this structure, will bypass be set to every other element.When implementing an above element, this system operates in the following manner.Data flow enters CODEC by data I/O port or image I/O port.Then, data flow is sent to buffer, is sent to multiplexer then.Data are sent to assigned memory or elementary section from multiplexer.Next element, for example the encryption element request be stored in the memory cell, by multiplexer and be sent to data in the encryption element.Then, encryption element can be carried out any in the multiple encryption technology.In case data are processed, these data just are sent to buffer, turn back to memory and particular memory location/section by multiplexer then.For receiving relevant all elements that flow to the control command of line operate to digital data, this processing will continue.Should be noted that each element is equipped with the address space of memory, so that retrieve according to initial order, this initial order is sent to overall processor controls from system processor, is sent to the modulator in the motion image chips then.At last, digital data stream is retrieved and is transmitted by image I/O port or FPDP from memory.The synchronizing signal of receiving according to CODEC or write instruction just takes place to send from the data of port.
Below will the element in the CODEC be described in more detail.Image I/O port is a bi-directional sample port.The synchronous transceive data of this port and synchronizing signal.The treatment element that should interlock is provided as known being used for of those of ordinary skills the frame of digital motion image stream is carried out pretreated several different methods.Preliminary treatment helps to make spatial vertical redundancy and the time field redundancy of showing up relevant.The time change element provides the filter of 9 taps, the wavelet transformation between these frame two ends, 9 tap filter implementation space.Filter can constitute and carry out convolution algorithm, and wherein, the termporal filter window slides between a plurality of frames two ends.Time change can comprise recursive operation, and this recursive operation can be realized many band (multi-band) time wavelet transformations, room and time combination and noise reduction filtering.Although the time change element can adopt example, in hardware to implement as digital signal processing integrated circuit, yet this element can constitute according to the metadata in the digital motion image stream (meta-data) or rely on system processor to receive coefficient value with memory filter.The same with the time change element, spatial transform element is implemented as digital signal processor, and this digital signal processor has the related memory cell that is used for the downloadable coefficients value.Spatial alternation in this preferred embodiment is the Symmetrical Two-dimentional acoustic convolver.This acoustic convolver has N tap position, and wherein, each tap all has the L coefficient of sampling/word basis cocycle (wherein, sampling or word can be defined as the position in groups).This spatial alternation can recurrence be carried out on input image data, so that carry out the multiband spatial wavelet transform, also can be used for the logical or such space filtering of noise reduction such as band.Entropy encoder/decoder element is being encoded between the piece two ends in a plurality of correlation times between the entire image two ends or temporarily.Entropy coder utilizes adaptive coder, and this adaptive coder is shown the minimum bit long symbols to the frequent data value table that takes place, and be the not too frequent value representation that takes place a length long symbols.Zero long haul distance (long run lengths) is represented as the single bit symbols of a plurality of null values in a plurality of byte information of representative.The more information of relevant entropy coder, referring to the 6th, 298, No. 160 United States Patent (USP)s, this United States Patent (USP) is transferred to the assignee identical with the present invention, and quotes in this article comprehensively, for your guidance.This CODEC also comprises encryption element, the encryption that this encryption element flows and the deciphering of stream.This CODEC can adopt Advanced Encryption Standard (AES) or other encryption technologies to implement.
Fig. 9 is the block diagram that the heterogeneous processing example in space is shown.In this embodiment, the mean data rate of digital motion image stream is 266MHz (4.23 lucky component/second).Each CODEC 920 all can handle with 66MHz speed, therefore, because required throughput is greater than the throughput of CODEC, thereby moving image is carried out heterogeneous processing.Digital motion image stream is sent in the DMR 910, and 910 pairs of each frames of this DMR are discerned, thereby this stream is divided into a plurality of space segments.This processing is undertaken by intelligent I/O port, and does not use the Digital Signal Processing element of DMR inside, so that adapt to the 266MHz bandwidth of image stream.The intelligent I/O port of exemplary DMR can be operated with the frequency speed of 533MHz, and the Digital Signal Processing element is operated with the maximum rate of 133MHz.When each frame during by segmentation, the intelligent I/O port of this DMR is sent to the image data stream of space segment in the frame buffer.Just as described above with reference to Figure 4, this CODEC sends signal to DMR, represents that it has prepared to receive data.DMR retrieves image data frame, and this image data frame is sent to a CODEC by intelligent I/O port.For among four CODEC each, this handles continuation, makes the 2nd CODEC receive second frame, and the 3rd CODEC receives the 3rd frame, and the 4th CODEC receives the 4th frame.This processing is recycled to a CODEC by returning, and is processed and be sent to memory cell from CODEC up to whole stream.In this example, this CODEC can carry out wavelet coding and compression and other motion image signal treatment technologies of frame.(definition motion image signal proceedings).
Figure 10 illustrates the block diagram that the spatial subbands of using DMR 1010 and CODEC 1020 is cut apart example.In this embodiment, Quad HD image stream (3840 * 2160 * 30 frame/second or 248MHz) is handled.Structure shown in a single day the input motion image stream enters just is divided into a plurality of colouring components frame by frame.The colouring component of frame adopts Y, Cb, Cr form 1030.The frame of 1010 pairs of image stream of this DMR carries out spatial manipulation, and each frequency band is sent to appropriate C ODEC is used for the time and handles.Because chromatic component only is that (Cb Cr), thereby only uses single DMR and two CODEC that each component is handled to half band.Luminance component (Y) at first comes time division multiplexing 1040 by the high rate multiplexer of operating with 248MHz, and wherein, even components is sent to a DMR 1110A, and the odd number component is sent to the 2nd DMR 1110B.Then, this DMR is used to export the two-dimensional convolution device of four frequency component L, H, V, D (low, high, vertical, diagonal angle).DMR carries out this task at average frame according to the speed of 64MHz.Be used for Cb is also used two-dimensional convolution device (having the filter factor different with the two-dimensional convolution device of Y component) with DMR 1010C, the D that the Cr component is handled, with the frequency division at the LH that obtains each component (low high) and VD (vertical diagonal angle).Then, the component of the frame of 1020 pairs of spatial division of this CODEC is handled.In this example, this CODEC carries out time change to a plurality of frames.(the additional announcement that needs relevant time change to handle).Should be appreciated that this DMR and CODEC are symmetrical fully, and can be used for image is carried out Code And Decode.
Although it will be understood by those skilled in the art that abovely to be described, yet can use digital motion image system chip to be used for decompression with regard to the compression aspect.This function is feasible, and this is because DMR can change by receiving different coefficient values with the interior element of CODEC, and under the situation of decompression, these elements can receive contrary coefficient.
In an alternate embodiment, scalable digital motion picture compressibility that is disclosed and method can be used as the computer program that uses with aforementioned calculation machine system and implement.This enforcement such as computer-readable medium (for example can be included in, disk, CD-ROM, ROM or fixed disk) fixing on such tangible medium, perhaps can instruct via modulator-demodulator or such as the series of computation machine that other such interface arrangements of the communication adapter that is connected with network by media are sent to computer system.This media can be tangible media (for example, light or analog communication line), the media that also can be to use wireless technology (for example, microwave, infrared or other transmission technologys) to implement.This series of computer instructions comprises above in this article with reference to the described all or part function of this system.It should be appreciated by those skilled in the art that these computer instructions can adopt the many programming languages that use with many Computer Architectures or operating system to write.And these instructions can be stored in such as in the such any storage device of semiconductor, magnetic, light or other storage devices, and can use such as light, infrared, microwave or the such any communication technology of other transmission technologys and transmit.What expect is, this computer program can be used as subsidiary have print or e-file (canned software), (for example be mounted with computer system in advance, on ROM of system or fixed disk) the packaged type medium issue, also can pass through network (for example, internet or World Wide Web (WWW)) issues from server or broadcasting bulletin system.Certainly, some embodiment of the present invention can be used as software (for example, computer program) and the combination of hardware is implemented.Other embodiment of the present invention can be embodied as whole hardware or whole software (for example, computer program).
Although above various one exemplary embodiment of the present invention have been done announcement, yet it will be understood by those skilled in the art that under the situation that does not deviate from true scope of the present invention, can carry out to realize the various changes and the modification of some advantage of the present invention.These and other obvious modifications will be contained by claims.
Claims (19)
1. scalable moving image compressibility that is used for digital motion image signal, wherein, this digital motion image signal has associated transmission rate, and this system comprises:
Decomposing module is used for receiving digital motion image signal according to transmission rate, this digital motion image signal is resolved into a plurality of components, and send these components according to this transmission rate; And
Compression module is used to receive each component from decomposing module, this component is compressed, and the component of this compression is sent to memory cell.
2. scalable moving image compressibility according to claim 1, wherein, this decomposing module comprises one or more resolving cells.
3. scalable moving image compressibility according to claim 1 wherein, compresses digital motion image signal according to transmission rate.
4. scalable moving image compressibility according to claim 1, this system further comprises programmable module, is used for the digital motion image signal after Route Selection between decomposing module and the compression module is decomposed.
5. scalable moving image compressibility according to claim 4, wherein, described programmable module is a field programmable gate array.
6. scalable moving image compressibility according to claim 5, wherein, described field programmable gate array is a Reprogrammable.
7. scalable moving image compressibility according to claim 1, wherein, described compression module comprises one or more compression units.
8. scalable moving image compressibility according to claim 7, wherein, the throughput of compression unit that multiply by the compression unit number is more than or equal to the transmission rate of digital motion image signal.
9. scalable moving image compressibility according to claim 7, wherein, the equal parallel work-flow of each compression unit.
10. scalable moving image compressibility according to claim 1, wherein, described decomposing module comprises one or more resolving cells.
11. scalable moving image compressibility according to claim 1, wherein, the equal parallel work-flow of described each resolving cell.
12. scalable moving image compressibility according to claim 1, wherein, described decomposing module is carried out the look decorrelation.
13. scalable moving image compressibility according to claim 1, wherein, described decomposing module is carried out the look rotation.
14. scalable moving image compressibility according to claim 1, wherein, the described decomposing module time of carrying out decomposes.
15. scalable moving image compressibility according to claim 1, wherein, described decomposing module is carried out spatial decomposition.
16. scalable moving image compressibility according to claim 1, wherein, described compression module uses sub-band coding.
17. scalable moving image compressibility according to claim 13, wherein, described sub-band coding uses small echo.
18. scalable moving image compressibility according to claim 1, wherein, described spatial decomposition is the heterogeneous decomposition in space.
19. a scalable system that is used for the digital motion picture input signal with associated transmission rate is carried out the moving image compression, this system comprises:
A plurality of compression blocks, each piece all has decomposing module and compression module;
Signal distributor, itself and compression blocks are coupled, and are used for the digital motion picture input signal is divided into a plurality of segmentations, and these a plurality of segmentations offer each compression unit to the different components of input signal;
This decomposing module resolves into a plurality of components with a segmentation, and sends these components; And
Compression module is used to receive the component from corresponding decomposing module, this component is compressed, and the component of this compression is sent to memory cell.
Applications Claiming Priority (6)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US26839001P | 2001-02-13 | 2001-02-13 | |
| US60/268,390 | 2001-02-13 | ||
| US28212701P | 2001-04-06 | 2001-04-06 | |
| US60/282,127 | 2001-04-06 | ||
| US35146302P | 2002-01-25 | 2002-01-25 | |
| US60/351,463 | 2002-01-25 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1547856A true CN1547856A (en) | 2004-11-17 |
Family
ID=27402067
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA028077148A Pending CN1547856A (en) | 2001-02-13 | 2002-02-13 | Scalable programmable motion image system |
Country Status (6)
| Country | Link |
|---|---|
| EP (1) | EP1362486A2 (en) |
| JP (1) | JP2005508100A (en) |
| KR (1) | KR20030081442A (en) |
| CN (1) | CN1547856A (en) |
| CA (1) | CA2438200A1 (en) |
| WO (1) | WO2002065785A2 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318534A (en) * | 2014-11-18 | 2015-01-28 | 中国电子科技集团公司第三研究所 | Real-time two-dimensional convolution digital filtering system |
| CN107024506A (en) * | 2017-03-09 | 2017-08-08 | 深圳市朗驰欣创科技股份有限公司 | A kind of pyrogenicity defect inspection method and system |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102006015392A1 (en) | 2006-04-03 | 2007-10-04 | Robert Bosch Gmbh | Driving and evaluating device for motor vehicle, has magnets with inner sections encompassing armature with slots, where slots are spaced apart in direction for receiving windings, over circumferential angle between specific ranges |
| JP4900470B2 (en) * | 2007-02-22 | 2012-03-21 | 富士通株式会社 | Moving picture coding apparatus and moving picture coding method |
| KR100958342B1 (en) * | 2008-10-14 | 2010-05-17 | 세종대학교산학협력단 | Method and apparatus for encoding and decoding video |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0314356B1 (en) * | 1987-10-27 | 1999-03-31 | Canon Kabushiki Kaisha | Predictive coding system |
| US5585852A (en) * | 1993-06-16 | 1996-12-17 | Intel Corporation | Processing video signals for scalable video playback using independently encoded component-plane bands |
| CA2134370A1 (en) * | 1993-11-04 | 1995-05-05 | Robert J. Gove | Video data formatter for a digital television system |
-
2002
- 2002-02-13 CN CNA028077148A patent/CN1547856A/en active Pending
- 2002-02-13 JP JP2002565366A patent/JP2005508100A/en active Pending
- 2002-02-13 WO PCT/US2002/004309 patent/WO2002065785A2/en not_active Application Discontinuation
- 2002-02-13 EP EP02713592A patent/EP1362486A2/en not_active Withdrawn
- 2002-02-13 KR KR10-2003-7010685A patent/KR20030081442A/en not_active Application Discontinuation
- 2002-02-13 CA CA002438200A patent/CA2438200A1/en not_active Abandoned
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318534A (en) * | 2014-11-18 | 2015-01-28 | 中国电子科技集团公司第三研究所 | Real-time two-dimensional convolution digital filtering system |
| CN104318534B (en) * | 2014-11-18 | 2017-06-06 | 中国电子科技集团公司第三研究所 | A kind of Real-time Two-dimensional convolutional digital filtering system |
| CN107024506A (en) * | 2017-03-09 | 2017-08-08 | 深圳市朗驰欣创科技股份有限公司 | A kind of pyrogenicity defect inspection method and system |
| CN107024506B (en) * | 2017-03-09 | 2020-06-26 | 深圳市朗驰欣创科技股份有限公司 | Pyrogenicity defect detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2005508100A (en) | 2005-03-24 |
| CA2438200A1 (en) | 2002-08-22 |
| KR20030081442A (en) | 2003-10-17 |
| WO2002065785A3 (en) | 2003-03-27 |
| WO2002065785A2 (en) | 2002-08-22 |
| EP1362486A2 (en) | 2003-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7076108B2 (en) | Apparatus and method for image/video compression using discrete wavelet transform | |
| US8559499B2 (en) | Frame buffer compression for video processing devices | |
| US20020141499A1 (en) | Scalable programmable motion image system | |
| EP1542476A2 (en) | Color space coding framework | |
| US9185423B2 (en) | Frame buffer compression for video processing devices | |
| CN1466853B (en) | Video processing method, device and system | |
| Jiang et al. | Scalable video coding using compressive sensing | |
| GB2617777A (en) | Temporal processing for video coding technology | |
| KR101285168B1 (en) | Transform domain sub-sampling for video transcoding | |
| US20160029024A1 (en) | Frame buffer compression for video processing devices | |
| JP2022536512A (en) | Encoder and method for encoding a sequence of frames | |
| EP1585338B1 (en) | Decoding method and decoding device | |
| US7848429B2 (en) | Hierarchical coding and decoding method | |
| CN1547856A (en) | Scalable programmable motion image system | |
| GB2488094A (en) | Image compression using sum and difference pixel replacement and lowest bit discarding | |
| EP1078504A2 (en) | Video compression with storage reduction, color rotation, combined signal and border filtering | |
| KR102247196B1 (en) | Vc-2 decoding using parallel decoding paths | |
| Kojima | LLVC—Low Latency Video Codec for Network Transfer | |
| AU2002245435A1 (en) | Scalable motion image system | |
| EP1652146A1 (en) | Implementation of the jpeg2000 compression algorithm in hardware | |
| Marino et al. | A DWT-based perceptually lossless color image compression architecture | |
| EP0848557A2 (en) | Subband image encoding method | |
| AU2017204642A1 (en) | Method, apparatus and system for encoding and decoding video data | |
| Christopoulou et al. | JPEG 2000 implementation on different computer platforms | |
| AU2017225027A1 (en) | Method, apparatus and system for encoding and decoding video data |
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 |