CN102480594B - Image processing apparatus of processing original high-resolution image data, image processing wafer and method thereof - Google Patents
Image processing apparatus of processing original high-resolution image data, image processing wafer and method thereof Download PDFInfo
- Publication number
- CN102480594B CN102480594B CN201010567706.8A CN201010567706A CN102480594B CN 102480594 B CN102480594 B CN 102480594B CN 201010567706 A CN201010567706 A CN 201010567706A CN 102480594 B CN102480594 B CN 102480594B
- Authority
- CN
- China
- Prior art keywords
- image data
- compression
- original high
- module
- memory
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 100
- 238000012545 processing Methods 0.000 title claims abstract description 44
- 238000007906 compression Methods 0.000 claims abstract description 125
- 230000006835 compression Effects 0.000 claims abstract description 125
- 230000015654 memory Effects 0.000 claims abstract description 59
- 239000011159 matrix material Substances 0.000 claims description 51
- 238000013139 quantization Methods 0.000 claims description 35
- 230000008569 process Effects 0.000 claims description 24
- 238000013144 data compression Methods 0.000 claims 6
- 238000005516 engineering process Methods 0.000 description 16
- 238000005070 sampling Methods 0.000 description 9
- 241001269238 Data Species 0.000 description 5
- 238000012546 transfer Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 230000000295 complement effect Effects 0.000 description 4
- 230000006837 decompression Effects 0.000 description 4
- 229910044991 metal oxide Inorganic materials 0.000 description 4
- 150000004706 metal oxides Chemical class 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 239000004065 semiconductor Substances 0.000 description 4
- 238000011002 quantification Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000005039 memory span Effects 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Abstract
The invention relates to a method for processing original high-resolution image data. The method comprises the following steps: obtaining the original high-resolution image data, compressing the original high-resolution image data and storing the compressed original high-resolution image data in a current frame area of a memory; acquiring the compressed original high-resolution image data from the current frame area of the memory and uncompressing the data; generating result image data according to the uncompressed original high-resolution image data and compression reference image data; storing the compressed result image data in a reference frame area of the memory.
Description
Technical field
The invention relates to a kind of image processor and image processing wafer and method of processing raw video data; A kind of particularly image processor and image processing wafer and method of processing original high parsing image data.
Background technology
The data volume little (about hundreds of kilobit groups or Kilobyte) that the picture producing in the dynamic video compression system of low image analytic degree in the past has.Because the not high storage space needing of resolution and the frequency range needs of picture are little, the picture that therefore each width need to store can be directly stored in memory.But along with the increase of current image analytic degree, each width picture also will get more and more to the demand of storage space and memory band width.
H.264, having in the market the very multiple standard for height parsing image data is carried out to encoding and decoding processing, is wherein that the height that is subject in current industry quite welcoming is resolved one of image encoding and decoding standard.Figure 1 shows that the calcspar of known image processing system, wherein known image processing system is to adopt H.264 standard that image data is encoded and decoded.As shown in Figure 1, known image processing system 10 comprises image sensor 20, memory 30 and encoder 40 H.264, and wherein memory 30 further comprises current picture district 31 and reference picture district 32.In addition, H.264 encoder 40 comprises comparison module 41, space modular converter 42, quantization modules 43 and coding module 44, is responsible for respectively image data coding different phase.
In the time carrying out dynamic video compression, the image sensor 20 of known image processing system 10 produces original high parsing image data H according to image, and also it is stored in the current picture district 31 of memory 30.In addition, the reference picture district 32 of memory 30 has one with reference to image data R.H.264 the comparison module 41 of encoder 40 will be obtained the original high image data H of parsing and with reference to image data and compare to judge the difference on image between two image datas from memory 30.
H.264 encoder 40 will produce quantification archives according to difference control space modular converter 42 and quantization modules 43 on above-mentioned image, produces data order code O for back-end processor processing for coding module 44 according to entropy compiling method or other compiling methods.In addition, H.264 encoder 40 is also controlled space modular converter 42, quantization modules 43 and other image processing modules and is reconstructed into the form that can be stored in memory 30 by quantizing archives, for the use of image data comparison next time.
But handled each the original height of known image processing system 10 is resolved image data H and all has sizable storage space (millions of bit groups or Megabyte) and corresponding stored device frequency range demand with reference to image data R.Thus, in the time carrying out dynamic video compression, resolve image data H and will need a large amount of storage space with reference to the storage of image data R for original height, wherein original height is resolved image data H and will take huge memory band width with reference to image data R.Above-mentioned dynamic video compression will increase the overall hardware cost of known image processing system 10 and lower the overall calculation efficiency of dynamic video compression.Therefore how in the situation that not increasing memory 30 space, process efficiently the high image data of resolving, real is one of important topic of current dynamic video compression.
Summary of the invention
The object of the present invention is to provide a kind of original high image processor and image processing wafer and method of resolving image data of processing, in order to save storage space and memory band width.
Image processing wafer of the present invention comprises compression module, memory and coding module, and image processing wafer is that self imaging inductor is obtained raw video data.Compression module comprises the first compression module and the second compression module, wherein the first compression module compression raw video data be stored in the current picture district of memory.In addition, the reference picture district of memory has with reference to image data, is wherein the raw video data of processing through coding before with reference to image data.
Coding module will be controlled respectively raw video data stored in the first compression module of compression module and the second compression module decompression memory and with reference to image data.Coding module will produce a result image data and by its compression and be stored in the reference picture district of memory according to difference between two image datas.
Compression module alternative of the present invention is compressed raw video data by undistorted compression technology or distortion compress technique and with reference to image data.When compression module is during with distortion compress technique reference compression image data, coding module can be by using recompile (Intra Refresh) or additive method encode to raw video data, with compensate because of before use the image that distortion compress technique produced to lose.In addition, the first compression module of different embodiment and the second compression module can be selected respectively suitable compression ratio to raw video data and compress with reference to image data in different compression ratios.
Accompanying drawing explanation
Figure 1 shows that the calcspar of known image processing system;
Described in Fig. 2, process the method step figure of original high parsing image data for the present invention;
Fig. 3 is that the present invention processes original high another embodiment that resolves image data method; And
Fig. 4 is that the present invention processes original high image processor calcspar of resolving image data.
Main element symbol description
100 image processors
110 image sensors
200 compression modules
210 first compression modules
220 second compression modules
300 memories
310 current picture districts
320 reference picture districts
400 coding modules
410 comparison modules
420 space modular converters
430 quantization modules
440 sequence generation modules
450 decoder modules
A raw video data
The raw video data of A ' through decompressing
The raw video data that B is compressed
C result image data
C ' is with reference to image data
The result image data that D is compressed
E transformed matrix
F quantization matrix
G encoded data sequence
Embodiment
The invention relates to a kind of image processing wafer and method of processing raw video data; Particularly a kind ofly process original high image processor and image processing wafer and the method for resolving image data that webcam module produces.Raw video data or coded image data storing before memory, are being compressed these raw video data or coded image data by the present invention, stores the required storage space of image data to save.
Image processing wafer of the present invention and method are preferably for webcam module; Webcam module is preferably by photo-sensitive cell or other inductors acquisition static image, dynamic image or film and converts it into the electronic installation that corresponding digital data are transmitted via network again.Webcam module can be independent device or is present in mobile phone, PC, electronic reader or various electronic installation.In preferred embodiment, webcam module can comprise for the video camera of PC or digital camera, but is not limited to this; In different embodiment, image processing wafer of the present invention and method also can be used for monitor, video camera or other are for converting pick-up image to the electronic product of corresponding digital data.
In addition, the original high image data of resolving according to webcam module is produced digital coded image data by image processing wafer of the present invention and method, wherein webcam module better by open world-wide web (Internet) by above-mentioned digital coded image transfer of data to plural terminal that world-wide web connects at least one of them, but be not limited to this.In different embodiments of the invention, webcam module also can be used among the enterprise network (Intranet) of relative closure and the inside terminals machine that digital coded image transfer of data is connected in enterprise network at least one of them.The network packet that webcam module can be used to carry out transfer of data is containing various cable networks interface and wireless network interface, to carry out the transmission of coded image data between the main frame in diverse location or device two.
Described in Fig. 2, process the method step figure of original high parsing image data for the present invention.As shown in Figure 2, this method comprises step S100, obtains raw video data.Step S100 uses the image sensor that comprises charge coupled device (charge-coupled device, CCD) to produce original high parsing image, but is not limited to this; In different embodiment; Image sensor also can comprise complementary metal oxide semiconductor (complementary metal oxide semiconductor, CMOS) or other and can produce according to image the image element of raw video data.
This method comprises step S110, compress original high resolve image data and by compressed raw video data storing in the current picture district of memory.Image sensor will obtained the compression module that transmits it to image processing wafer after raw video data, and wherein compression module will be stored to the current picture district of memory after compression raw video data.In the present embodiment, compression module adopts the undistorted compression technologies such as entropy compiling method to process raw video data, to preserve the integrality of raw video data but to be not limited to this; In different embodiment, compression module also can be according to distortion compression techniques process raw video data such as colourity sampling.
As mentioned above, this method was first being carried out image compression to save thus the capacity of memory to it by raw video data storing before memory.Thus, even if the data that image sensor produces are raw video data of corresponding high-res, original memory also can be in order to store more than raw video data in the past.
In the present embodiment, step S110 compresses the original high image data of resolving according to fixing compression ratio, but is not limited to this; In different embodiment, step S110 alternative is carried out selectivity according to original high size of data of resolving image data or other conditions and is selected one of them to compress in plural compression ratio.
In addition, the memory that the present embodiment is practical is preferably dynamic random access memory (Dynamicrandom access memory, DRAM), but is not limited to this; Memory used in the present invention also can comprise static RAM (Static random access memory, etc. SRAM) volatile storage or the electronics formula of erasing can be made carbon copies the non-volatility memorizers such as read-only memory (Electrically-Erasable Programmable Read-Only Memory, EEPROM).
This method comprises step S120, obtains compressed original high parsing image data and also decompresses with reference to image data from current picture district and the reference picture district of memory.In the present embodiment, memory comprises one for storing the reference picture district of plural reference image data, and wherein above-mentioned is to process original high result of resolving after image data before with reference to image data.Image processing wafer comprises a comparison module, in order to resolve image data and to produce a result image data with reference to image data according to the original high of above-mentioned decompression.
The present invention processes original high method of resolving image data and comprises step S131, resolves image data and calculates image difference with reference to image data according to the original height decompressing.It is the original high image data of resolving of plural number producing for the treatment of photography that the present embodiment is processed original high method of resolving image data.Because photography has continuity between captured, therefore original high parsing only has the difference on partial image conventionally between image data and the reference image data that calculated before.Therefore step S131 calculates the difference of image against image difference, has saved the calculation procedure of image processing wafer and has calculated required hardware resource.
The present invention processes original high method of resolving image data and further comprises step S132, according to the image difference image data that bears results.As mentioned above, step S131 moves and obtains two image differences between image data by the raw video data to decompressing obtained in step S120 and with reference to image data to compare.Step S132, comprises and above-mentioned image difference will be changed into a transformed matrix, and wherein the transformed matrix of the present embodiment is the matrix of a 8X8, but is not limited to this; In different embodiment, transformed matrix also comprises 4x4 matrix.Initial data shelves comprise plural number of sampling certificate, and wherein space modular converter is according to converting the information such as brightness (Luminance), colourity (Chrominance) and concentration (Chroma) to by each number of sampling.In addition, at number of sampling, according to after converting, space modular converter converts the brightness of each number of sampling certificate, chroma data respectively to the transformed matrix of one 8X8, but is not limited to this; In different embodiment, transformed matrix also comprises 4x4 matrix, and the data that wherein above-mentioned transformed matrix comprises are to be arranged in spatial domain.
Step S132 comprises above-mentioned transformed matrix is quantized and is translated into a quantization matrix.Because human eye is more insensitive for the brightness variation of image, therefore the feature that quantization modules is intended to according to above-mentioned human eye, brightness be changed goes to reduce the data volume in transformed matrix with amplitude coefficient greatly.Quantization modules by the coefficient in transformed matrix divided by a constant, and after by round off (Round Off) by this coefficient adjustment to nearest integer.In the present embodiment, after most of common quantization step of coefficient originally with higher amplitude, will be adjusted to 0, therefore the quantification of the present embodiment action not only reduces the data volume that transformed matrix comprises, also save the required space of memory storage data simultaneously.
Step S132 is carried out the action of inverse quantization (Inverse Quatization) and inverse transform (Inverse Transform) and quantization matrix is converted to and has the reference image data that is same as raw video data format after being included in producing qunantization matrix.
This method will further perform step S140 after image data bearing results, and reference compression image data also deposits compressed result image data in the reference picture district of memory.Coding module is the compression module that transfers to image processing wafer with reference to image data, and compression module will be stored in the reference picture district of memory after compression result image data, use during once with reference to comparison between image data and raw video data on carrying out for step S131.In other words, the new result image data producing is using the reference image data using in step S131 next time.
As shown in Figure 2, the present invention processes original high method of resolving image data and comprises step S140, compression result image data also deposits compressed result image data in the reference picture district of memory, is used for more next original high basis of resolving image data for the present invention.In addition, the step S140 of the present embodiment adopts the undistorted compression technology compression result image datas such as entropy compiling method, but is not limited to this; In different embodiment, compression module also can be according to quantizing the distortion compress technique compression result image datas such as compression method.In addition, the step S140 of the present embodiment compresses result image data according to fixing compression ratio, but is not limited to this; In different embodiment, step S140 is also alternative selects one of them to compress result image data in plural compression ratio.
In the embodiment shown in Figure 2, the present invention processes original high method of resolving image data will compress respectively raw video data and result image data in step S110 and step S140, but be not limited to this;
In different embodiment, the present invention process original high method of resolving image data also alternative only compress raw video data and result image data one of them in the situation that raw video data and result image data are stored to memory.
Figure 3 shows that the present invention processes another embodiment of original high method of resolving image data.In the present embodiment, step S140 alternative is carried out compression result image data by undistorted compression technology or distortion compress technique.Step S140 is better carrys out compression result image data with undistorted compression technology, with the integrality of saving result image data, but is not limited to this; Step S140 also can carry out compression result image data by distortion compress technique, further to reduce the data volume of result image data and to save the required storage space of memory.
As shown in Figure 3, the present invention processes original high method of resolving image data and comprises step S133, judges the technology of compression result image data.When result image data is while compressing with undistorted compression technology, compressed result image data will directly be stored to the reference picture district of memory.But when result image data is while compressing with distortion compress technique, compressed result image data will be lost and have image infringement because of partial data.
In order to compensate the image infringement that adopts distortion compress technique and produce, the present invention processes original high method of resolving image data and further comprises step S134, according to the compress mode selectivity of result image data, result image data is carried out to image reparation.Because incomplete result image data will be subject to further infringement because of distortion compress technique, resolve image data and calculate image difference with reference to image data according to the original height decompressing for step S131.As the step S140 of the present embodiment adopts distortion compress technique to carry out compression result image data, step S134 produces repairing result image data image infringement because using distortion compress technique, to avoid step S140 to continue produce impaired reference image data and be stored to memory because above-mentioned image damages.As can be seen here, step S134 can avoid further being transmitted and expanding because of the suffered infringement of result image data.
In the present embodiment, step S134 repairs by Intra recompile technology the infringement that result image data produces because of distortion compress technique by controlling comparison module.Comparison module, by selecting one of them in the plural reference image data that periodically self-reference picture district has and being alternatively to the result image data that needs repairing, expands with the infringement of avoiding result image data, but is not limited to this.In different embodiment, comparison module also can picture segmentation (Picture Segmentation) or other image repairing techniques the prejudicial part of result image data tool is repaired.
Figure 4 shows that the present invention is for the treatment of original high image processor 100 calcspars of resolving image data.Image processor 100 comprises interconnective image sensor 110, compression module 200, memory 300 and coding module 400.As shown in Figure 4, compression module 200 comprises the first compression module 210 and the second compression module 220, and memory 300 comprises current picture district 310 and reference picture district 320.The first compression module 210 is to connect image sensor 110, current picture district 310 and coding module 400 simultaneously.The second 220 of compression modules are to connect reference picture district 320 and coding module 400 simultaneously.In addition, coding module 400 comprises comparison module 410, space modular converter 420, quantization modules 430, sequence generation module 440 and rebuilds module 450.
In the embodiment shown in fig. 4, image sensor 110 will be converted into the video camera of raw video data A or other and can be used for the image sensor 110 of digital camera or other imaging devices for becoming known for according to image.The image sensor 110 of the present embodiment can produce the raw video data A of above (for example 1920x1080) resolution of corresponding 1280*720, but is not limited to this; In different embodiment, image sensor 110 is the alternative raw video data A that produces the following resolution of corresponding 1280*720 also.In addition, above-mentioned raw video data A comprises .raw, the raw video data A that .yuv or other industries are known.In addition, the image sensor 110 of the present embodiment comprises charge coupled device (charge-coupled device, CCD), but is not limited to this; In different embodiment; Image sensor 110 also can comprise complementary metal oxide semiconductor (complementary metal oxide semiconductor, CMOS) or other and can produce according to image the image element of raw video data A.
Image sensor 110 will transfer to the first compression module 210 raw video data A, raw video data A be compressed according to image compression technology for the first compression module 210.After the first compression module 210, compressed raw video data B will be stored in to the current picture district 310 of memory 300.Have compared with small data quantity afterwards because compressed raw video data B is compressed, the first compression module 210 is to have been saved and stored the required space of raw video data A by image compression technology as can be seen here.In addition, first compression module 210 of the present embodiment is to process raw video data A according to undistorted compression technologies such as entropy compiling methods, but is not limited to this; In different embodiment, the first compression module 210 also can be according to quantizing the distortion compression techniques process raw video data A such as compression method.
In the present embodiment, there is compressed result image data D in the reference picture district 320 of memory 300.Coding module 400 will compare with reference to image data C ' and after decompressing the difference between raw video data A ' will produce new result image data C according to this difference.Encode because above-mentioned coding module 400 only needs two differences between image, therefore do not need whole raw video data A to be converted into result image data C and to be stored in reference picture district 320.Thus, coding module 400 can be saved the required memory of memory 300 300 capacity by compressing image data.
Carry out image data relatively before, the raw video data A ' that the comparison module of coding module 400 410 needs from picture district 310 obtain after decompression at present.The first compression module 210 will be obtained compressed raw video data B from current picture district 310 and to its action that decompresses.In addition, the second compression module 220 is obtained compressed result image data D by the reference picture district 320 from memory 300 and it is decompressed and is moved to obtain with reference to image data C '.Will be simultaneously after comparison module 410 from two compression modules 210,220 obtain raw video data A ' through decompressing and with reference to image data C ' and relatively two image datas difference between corresponding image.But in different embodiment, coding module 400 also can directly produce a result image data C according to the raw video data A ' after decompressing.
The space modular converter 420 of coding module 400 will convert a transformed matrix E to above-mentioned image difference, and wherein the transformed matrix E of the present embodiment is the matrix of a 8x8, but is not limited to this; In different embodiment, transformed matrix also comprises 4x4 matrix.Image difference comprises plural number of sampling certificate, and wherein space modular converter is according to converting the information such as brightness (Luminance), colourity (Chrominance) to by each number of sampling.In addition, at number of sampling, according to after converting, space modular converter 420 converts the brightness of each number of sampling certificate, chroma data respectively to the transformed matrix E of 8X8 or 4x4, and the data that wherein above-mentioned transformed matrix E comprises are to be arranged in a spatial domain.
In addition, the space modular converter 420 of coding module 400 is for above-mentioned transformed matrix E is changed to frequency domain from spatial domain.Originally the coefficient that transformed matrix E comprises is all representing brightness, the colourity of image a certain position in space.Space modular converter 420 changes into each coefficient of transformed matrix E the amplitude of a frequency spectrum in frequency domain.The space modular converter 420 of the present embodiment uses discrete cosine transform to carry out the conversion in above-mentioned space, but is not limited to this; In different embodiment, space modular converter 420 also can comprise wavelet transformation, Fourier transform or other are for changing the method to frequency domain by signal from spatial domain.
It is that the transformed matrix E that the quantization table for comprising according to memory 300 is exported space modular converter 420 converts a quantization matrix F to that coding module 400 further comprises quantization modules 430.Because human eye is more insensitive for the brightness variation of image, therefore the feature that quantization modules 430 is intended to according to above-mentioned human eye, brightness be changed goes to reduce the data volume in transformed matrix E with amplitude coefficient greatly.Quantization modules 430 by the coefficient in transformed matrix E divided by a constant, and after by round off (Round Off) by this coefficient adjustment to nearest integer.In the present embodiment, after most of common quantization step of coefficient originally with higher amplitude, will be adjusted to 0, therefore the quantification of the present embodiment action has reduced the data volume that transformed matrix comprises and stored afterwards needed storage space.
As shown in Figure 4, coding module 400 comprises sequence generation module 440 and rebuilds module 450, and wherein sequence generation module 440 is for quantization matrix F being converted to an encoded data sequence G.In addition, the encoded data sequence G of the present embodiment has the data format identical with initial data code and same being stored in memory 300.In the present embodiment, because DC coefficient that quantization matrix F comprises etc. has coefficient of different nature, therefore the sequence generation module of the present embodiment produces coding schedule according to the kind of coefficient with different coding methods, and changes quantization matrix F according to coding schedule afterwards.In the present embodiment, sequence generation module 440 use huffman coding methods and length coding method are encoded to coefficients different in quantization matrix F, but are not limited to this; In different embodiment, sequence generation module 440 also can encode to produce encoded data sequence G to coefficients different in quantization matrix F with Arithmetic Expressions Using coding or entropy compiling method.
In the embodiment shown in fig. 4, the quantization matrix F that quantization modules 430 produces also can carry out re-quantization processing (Inverse-Quantization) and produce transformed matrix E according to original quantization table, and the coefficient (coefficient that frequency is higher) after wherein quantizing in transformed matrix E reverts back in fact the state before being quantized.Process through re-quantization the transformed matrix E producing and will be received and transmit digital signal by space modular converter 420 to rebuilding module 450 after transition matrix E is carried out to inverse transform, rebuild the digital signal of receiving with the image data C that bears results for rebuilding module 450.
In addition, second compression module 220 of the present embodiment carries out own coding module 400 reception result image data C selectivity image compression processing and compressed result image data D is stored to reference picture district 320 result image data C with undistorted compression technology or distortion compress technique.
In the embodiment shown in fig. 4, the first compression module 210 and the second compression module 220 are according to fixing compression ratio, raw video data A and result image data C to be compressed, but are not limited to this; In different embodiment, the first compression module 210 and the second compression module 220 also can have according to setting the compression ratio of plurality of distinct, and what wherein user can be according to memory sets the first compression module 210 or the different compression ratios of the second compression module 220 use are carried out the compression of image data by the condition such as state, memory span.
In addition, in the present embodiment, the first compression module 210 alternatives are compressed raw video data A by undistorted compression technology or distortion compress technique.Similarly, the second compression module 220 is to use the undistorted compression technologies such as huffman coding, arithmetic coding or other to process with reference to image data, but is not limited to this; The second compression module 220 can carry out result image data C by distortion compress techniques such as wavelet coding, frequency division codings in order to save storage space.But distortion compress technique will make result image data C loss partial data.In other words, distortion compress technique will cause image infringement to result image data C.
In the time that the second compression module 220 uses distortion compress technique compression result image data C and it is caused to image infringement, the comparison module 410 use Intra recompile technology of the present embodiment are repaired the infringement producing because of distortion compress technique with reference to image data.Thus, comparison module 410 can be avoided above-mentioned image infringement to exist and therefore make image processing afterwards continue to produce incomplete with reference to image data.As can be seen here, comparison module 410 can be avoided because further being transmitted and expand with reference to the suffered infringement of image data by repairing.Comparison module 410 will be selected one of them in the plural reference image data that periodically self-reference picture district 320 has and will be alternatively to the reference image data that must repair, to avoid the infringement expansion with reference to image data, but be not limited to this.In different embodiment, comparison module 410 also can picture segmentation (Picture Segmentation) or other image repairing techniques to repairing with reference to the prejudicial part of image data tool.
In the embodiment shown in fig. 4, the present invention compresses with the first compression module 210 and the second compression module 220 the result image data C that raw video data A and coding module 400 produce respectively for the treatment of original high image processor 100 of resolving image data, but is not limited to this.In different embodiment, in order to save the required time of compression module 200 deal with data, compression module 200 also can only compress raw video data A or only carry out compression result image data C with the second compression module 220 with the first compression module 210.In other words,, in different embodiment, only by compressing raw video data A and result image data C, one of them also saves corresponding decompression step to compression module 200 thus.
Although aforesaid description and diagram have disclosed preferred embodiment of the present invention, must recognize variously increase, many modifications and replace and may be used in preferred embodiment of the present invention, and can not depart from spirit and the scope of the principle of the invention defining as appended claims.Those of ordinary skills can know from experience the present invention and may be used in the modification of a lot of forms, structure, layout, ratio, material, element and assembly.Therefore, the embodiment and all viewpoints that disclose at this herein, should be regarded as illustrating the present invention, but not in order to limit the present invention.Scope of the present invention should be defined by appended claim book, and contains its legal equivalents, is not limited to previous description.
Claims (7)
1. process a webcam module and produce an original high method of resolving image data, comprise the following step:
Obtain this original high image data of resolving;
Obtain one with reference to image data from a memory;
Produce a result image data according to this original high parsing image data and this with reference to image data; And
Compress this original high resolve image data and this result image data at least one of them and highly resolve image data or this compressed result image data is stored to this memory by original compressed this;
Wherein this original high image data compression step of resolving comprises according to a distortion compress technique and compresses this original high image data of resolving, this result image data compression step comprises according to this distortion compress technique and compresses this result image data, when this result image data is during with the compression of this distortion compress technique, the method comprises:
This result image data is carried out image processing and produces a transformed matrix;
Convert this transformed matrix to a quantization matrix according to a quantization table; And
Convert this quantization matrix to an encoded data sequence according to a coding schedule.
2. the method for claim 1, wherein this original high image data compression step of resolving comprises a current picture district that this compressed original high parsing image data is stored to this memory, and this result image data compression step comprises a reference picture district that this compressed result image data is stored to this memory.
3. the method for claim 1, wherein this original high image data compression step of resolving comprises according to one first compression ratio and compresses this original high image data of resolving, and this result image data compression step compresses this result image data according to one second compression ratio.
4. process an original high image processing wafer of resolving image data, for a webcam module, this image processing wafer comprises:
One memory, comprises one with reference to image data;
One coding module, produces a result image data according to this original high parsing image data and this with reference to image data; And
One compression module, compress this original high resolve image data and this result image data at least one of them and highly resolve image data or this compressed result image data is stored to this memory by original compressed this;
Wherein this compression module compresses this original high image data of resolving according to a distortion compress technique, this compression module compresses this result image data according to this distortion compress technique, in the time that this compression module compresses this result image data according to this distortion compress technique, this coding module will carry out image processing and produce a transformed matrix this original high image data of resolving after decompressing, and this coding module converts this transformed matrix one quantization matrix to and converts this quantization matrix to an encoded data sequence according to a coding schedule according to a quantization table.
5. image processing wafer as claimed in claim 4, wherein this compression module comprises one first compression module and one second compression module, this first compression module compresses this original high parsing image data and is stored to this memory, and this second compression module compresses this result image data and is stored to this memory.
6. image processing wafer as claimed in claim 4, wherein this memory comprises current picture district and a reference picture district, respectively in order to store original high image data and this compressed result image data of resolving of compressed this, this coding module respectively from this current picture district and this reference picture district obtain this through decompressing original high resolve image data and this with reference to image data to produce this result image data.
7. image processing wafer as claimed in claim 4, wherein this compression module comprises at least one the first compression ratio and one second compression ratio, this compression module compresses this original high image data of resolving according to this first compression ratio, and this compression module compresses this result image data according to this second compression ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010567706.8A CN102480594B (en) | 2010-11-23 | 2010-11-23 | Image processing apparatus of processing original high-resolution image data, image processing wafer and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010567706.8A CN102480594B (en) | 2010-11-23 | 2010-11-23 | Image processing apparatus of processing original high-resolution image data, image processing wafer and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102480594A CN102480594A (en) | 2012-05-30 |
| CN102480594B true CN102480594B (en) | 2014-05-28 |
Family
ID=46093062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010567706.8A Active CN102480594B (en) | 2010-11-23 | 2010-11-23 | Image processing apparatus of processing original high-resolution image data, image processing wafer and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102480594B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6389160B1 (en) * | 1999-04-09 | 2002-05-14 | Picsurf, Inc. | Hybrid wavelet and JPEG system and method for compression of color images |
| CN1976475A (en) * | 2005-11-14 | 2007-06-06 | 联发科技股份有限公司 | Image processing apparatus and method |
| CN101426139A (en) * | 2008-12-08 | 2009-05-06 | 中国科学院长春光学精密机械与物理研究所 | Image compression apparatus |
| CN101835038A (en) * | 2009-03-13 | 2010-09-15 | 智微科技股份有限公司 | Image compressing method and related device thereof |
-
2010
- 2010-11-23 CN CN201010567706.8A patent/CN102480594B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6389160B1 (en) * | 1999-04-09 | 2002-05-14 | Picsurf, Inc. | Hybrid wavelet and JPEG system and method for compression of color images |
| CN1976475A (en) * | 2005-11-14 | 2007-06-06 | 联发科技股份有限公司 | Image processing apparatus and method |
| CN101426139A (en) * | 2008-12-08 | 2009-05-06 | 中国科学院长春光学精密机械与物理研究所 | Image compression apparatus |
| CN101835038A (en) * | 2009-03-13 | 2010-09-15 | 智微科技股份有限公司 | Image compressing method and related device thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102480594A (en) | 2012-05-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2487473C2 (en) | Switching between discrete cosine transform coefficient coding modes | |
| JP6710755B2 (en) | Entropy coding technique for non-4:4:4 chroma subsampling display stream compression (DSC) | |
| KR101890203B1 (en) | Compression and decompression of reference images in a video encoder | |
| RU2503138C2 (en) | Embedded graphics coding for images with sparse histograms | |
| EP0580454A2 (en) | Coding and decoding of digital data | |
| EP3354031A1 (en) | Video decoder conformance for high dynamic range (hdr) video coding using a core video standard | |
| KR20170016849A (en) | Entropy coding techniques for display stream compression (dsc) | |
| CN1366778A (en) | Video compression | |
| CN102823245A (en) | Video transmission system having reduced memory requirements | |
| CN100553151C (en) | The equipment and the method that are used for multiple description coding | |
| US5844607A (en) | Method and apparatus for scene change detection in digital video compression | |
| CN104270641A (en) | Method for processing transformation coefficients and device | |
| KR100359821B1 (en) | Method, Apparatus And Decoder For Motion Compensation Adaptive Image Re-compression | |
| JP2880051B2 (en) | Video signal encoding device using 1D / 2D DCT | |
| WO2009031904A2 (en) | Method for alternating entropy coding | |
| US20230222696A1 (en) | Image processing method and apparatus, device, and computer-readable storage medium | |
| US20090034634A1 (en) | Differential coding with lossy embedded compression | |
| CN102480594B (en) | Image processing apparatus of processing original high-resolution image data, image processing wafer and method thereof | |
| WO2003096700A1 (en) | Method for decoding video encoded as compressed bitstream encoded as a plurality of blocks | |
| KR100219133B1 (en) | Method and apparatus for selecting transform coefficient of transform encoding system | |
| TWI436649B (en) | Image processing device, image processing chip and method for processing raw high resolution image data | |
| US11388445B1 (en) | Mosquito noise smoothing between different video subsections encoded with different compression methods within a video frame | |
| JP2000032458A (en) | Image compression method | |
| CN116980609A (en) | Video data processing method, device, storage medium and equipment | |
| KR100744442B1 (en) | Improved cascaded compression method and system for digital video and images |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180129 Address after: Taiwan Hsinchu County Tai Yuan Street China jhubei City, No. 26 4 floor 1 Patentee after: MStar Semiconductor Co., Ltd. Address before: County Road nine Taiwan 302 Hsinchu County China jhubei City, No. 145 5 floor Patentee before: Alpha Imaging Technology Corp. |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20180315 Address after: Taiwan Hsinchu County Chinese jhubei City, Taiwan two yuan Street No. 1 10 Floor 2 Patentee after: Juxing electronic Limited by Share Ltd Address before: Taiwan Hsinchu County Tai Yuan Street China jhubei City, No. 26 4 floor 1 Patentee before: MStar Semiconductor Co., Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190115 Address after: Area 1501, Innovation Building A, Software Park, Xiamen Torch High-tech Zone, Xiamen City, Fujian Province Patentee after: Xiamen Xingchen Technology Co., Ltd. Address before: 2 on the 10th floor, 1 Taiyuan Second Street, Zhubei City, Hsinchu County, Taiwan, China Patentee before: Juxing electronic Limited by Share Ltd |
|
| TR01 | Transfer of patent right | ||
| CP03 | Change of name, title or address |
Address after: 361005 1501, zone a, innovation building, software park, Xiamen Torch High tech Zone, Xiamen, Fujian Patentee after: Xingchen Technology Co.,Ltd. Address before: 361006 1501, zone a, innovation building, software park, Xiamen Torch High tech Zone, Xiamen, Fujian Patentee before: Xiamen Xingchen Technology Co.,Ltd. |
|
| CP03 | Change of name, title or address |