CN100477789C - A method of inter-frame reinforcement based on Moving Joint Picture Experts Group 2000 - Google Patents
A method of inter-frame reinforcement based on Moving Joint Picture Experts Group 2000 Download PDFInfo
- Publication number
- CN100477789C CN100477789C CN 03149199 CN03149199A CN100477789C CN 100477789 C CN100477789 C CN 100477789C CN 03149199 CN03149199 CN 03149199 CN 03149199 A CN03149199 A CN 03149199A CN 100477789 C CN100477789 C CN 100477789C
- Authority
- CN
- China
- Prior art keywords
- data
- frame
- loss
- reference frame
- image
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention discloses a method of intensifying video frequency image. It includes the following procedures: set up image reference pins that are used in intensifying the compensation for the video frequency image pins whose data is lost; divide the pins of loss of data into data blocks that are MxN pixel(both M and N are integer); form the motion estimation between reference pins and pins of loss of data, so it's easy to search the data block that is matching the data block whose data is lost in reference pins; make use of the above pins to intensify the lost data blocks in the pins of loss of data. The method can use the relevance between image pins to compensate the loss of image data caused by network transmission congestion, and upgrade the image quality.
Description
Technical field
The present invention relates to a kind of video image interframe Enhancement Method, particularly a kind of video image based on motion JPEG (joint photographic experts group) 2000 (Motion JPEG 2000) compression is carried out the method that interframe strengthens, this method can be recovered the details of the video image lost as much as possible at receiving terminal.
Background technology
Along with the popularizing of multimedia application, the development of digital video technology and increasing of network epigraph transmission, the for example use of video telephone, the remote browse of image and retrieval, global multi-user virtual environment are shared etc., and the research of image processing techniques is become more and more important.
Treatment of picture comprises image compression, figure image intensifying and recovery, and image removes makes an uproar and description, classification and the identification of image.For IMAQ, storage and transmission course, press for fast-developing Image Compression.The data volume that one width of cloth has colored (24bit/ pixel) digital picture of intermediate resolution (640*480) is about 7.37Mbit, and huge data volume has brought very big difficulty for the storage and the transmission of image.Therefore, before transmission, must encode to image.
Therefore traditional JPEG compress technique can't satisfy the requirement of people to the multimedia image data, needs a kind of with better function, Standard of image compression that efficient is more remarkable.JPEG 2000 arises at the historic moment under this background.
Motion JPEG (joint photographic experts group) 2000 (Motion JPEG 2000) is a kind of inner frame coding method to moving image.It does not utilize frame-to-frame correlation, and the compression means that is to use JPEG 2000 to provide is removed the interior redundancy of the frame of each frame in the video sequence frame by frame.
As time goes on, people are more and more higher to the requirement of image Compression.In addition, more and more general based on the video transmission of the Internet and wireless channel.In actual applications, some important images are arranged, need transmit by the Internet as satellite remote sensing images, medical image etc.And these image transmission requests are guaranteed higher picture quality.
Yet owing to be subjected to the restriction of channel width, the video transmission of being undertaken by the Internet or wireless channel is subjected to the influence of network congestion usually.When network occurs when congested, router blocks packets of some expression details selectively, and this will cause the reduction of the picture quality of the intercepted frame of information.
In JPEG 2000, the spatial domain signal removes direct current through changing such as the color space, after the pre-treatment of normalization and so on, carries out wavelet transform (DWT), to remove the correlation between the data.The schematic diagram that the one-dimensional discrete wavelet transformation decomposes has been shown among Fig. 1.The basic filtering mode of one dimension DWT is by the high and low frequency bank of filters, and primary signal is decomposed on the frequency domain two mutually not on the subband of aliasings, and by the down-sampled invariable number that comes inhibit signal.Remaining low frequency signal has correlation toward contact behind the DWT.Can do k (k is an integer) level sub-band division thereon again, till correlation between signals can be ignored.
Two dimension DWT then is on the capable DWT of one dimension and down-sampled basis, again column signal is carried out DWT one time, and carries out down-sampled with identical before the data volume after the assurance conversion and the conversion.Through after the two-dimentional DWT conversion once, original image is divided into four subbands.A wherein unique LL subband (row low frequency and row low frequency) still can continue to decompose, through forming hierarchy after the 3-5 level.This hierarchy makes JPEG 2000 support the code flow structure of multiresolution.
Wavelet coefficient after the DWT need be through quantification, and a special character of the quantizer of JPEG 2000 has been to introduce the notion of " Dead Zone ".For the quantizer of each subband, the width of " DeadZone " all is 2 times of other step-length.This means that if adopt step-length in subband b be Δ
bQuantizer, then (i j) can use this subband coefficient y
As the quantization index value, wherein sign () represents the symbol of a variable,
Expression rounds downwards.The advantage of this quantification manner is to provide a kind of progressive bit rate mode according to the signal to noise ratio requirement, promptly can quantize signal with smaller step size earlier.Carry out inverse quantization according to user's different demands with the step-length of broad 2 integral number power of original step-length (generally be doubly) then.At this moment, some least significant bits (LSB) that only need abandon the quantization index value get final product.
Need then (EBCOT) the embedded block encoding (Embedded Block Coding with Optimal Truncation) that is optimized point of cut-off through the wavelet coefficient that quantizes.EBCOT is a kind of coded system based on bit plane (bit plain).According to plane order from high to low, extract the data that are in the identical bits plane and encode.So just provide picture quality by thick gradual coding to essence.Simultaneously, EBCOT also provides the support to many point of cut-offs,, in order to adapt to certain compression bit rate, can block this code stream at any one point of cut-off of JPEG 2000 code streams that is.Be equivalent to abandon all inapparent coefficients in the present bit plane and all remarkable coefficients than low order.
Code stream through EBCOT is sent into arithmetic encoder, is packaged into Motion JPEG 2000 (MJP2) code stream, is sent to decoding end.The decoder of decoding end according to top described opposite order, that is, carry out arithmetic decoding, contrary EBCOT, re-quantization, inverse discrete wavelet transform and reprocessing obtain decoded each two field picture.
Progressive telescopic decoding schema is for the transmission no less important of video.In the Internet or wireless channel, owing to be subjected to the restriction of channel width, video transmission often is subjected to the influence of network congestion, occurs phenomenons such as picture pause when causing online watching.If adopt incremental model transmission MJP2 code stream, when running into network congestion, router can initiatively block the code stream of each frame, guarantees the unimpeded of network to reduce data volume.At this moment, the user still can watch continuous picture, but the detailed information of relevant image lost, and this will inevitably cause decrease in image quality.
Therefore, need a kind of method that recover the detailed information of the image lost in decoding end as much as possible.
Summary of the invention
The purpose of this invention is to provide video image that a kind of enhancing receives to improve the method for picture quality, this method can utilize the frame-to-frame correlation of each frame of moving image that decoding end receives to recover to be subjected to the details of the LOF of information loss to a certain extent, thereby strengthens picture quality.
In order to realize purpose of the present invention, a kind of video image enhancing method is provided, comprise step: set up the image reference frame that is used for the video frame image that data are lost is strengthened compensation; The loss of data frame is divided into size is the data block of M * N pixel; Utilize inverse discrete wavelet transform to convert the frequency-region signal of reference frame to the spatial domain signal;
At the x axle, the direction of y axle and x=y moves a pixel with resulting described spatial domain signal, and the picture signal of resulting three moving directions is carried out wavelet transform to obtain the frequency-region signal of reference frame through displacement; In the estimation of between reference frame and the loss of data frame of displacement, taking exercises, so that in reference frame, occur the data block that the data block of loss of data is mated in search and the loss of data frame; Be used in the matched data piece that searches in the reference frame and strengthen the data block of losing in the loss of data frame.
Adopt moving image Enhancement Method of the present invention can utilize the correlation between the picture frame to compensate enhancing, improve picture quality because of the loss of image data that the congested grade of Network Transmission causes.
Description of drawings
By below in conjunction with the detailed description of accompanying drawing to the embodiment of the invention, to make above-mentioned and purpose, feature and advantages others of the present invention become more obvious and be more readily understood, need to prove that these embodiment only are for the present invention is described, rather than technical scheme of the present invention is limited to this, wherein:
Fig. 1 is the schematic diagram of the multistage wavelet transform of explanation;
Fig. 2 is enhancement processing method is carried out in explanation to video image according to embodiments of the invention a flow chart;
Fig. 3 shows employing image enchancing method according to the present invention carries out the Y-PSNR gain of the image after the enhancement process to image schematic diagram.
Embodiment
Describe the embodiment that the frame-to-frame correlation of utilizing image according to the present invention strengthens the method for image in detail below with reference to accompanying drawing.
At first, introduce JPEG 2000 standards and wavelet transform.JPEG 2000 has following characteristics as the upgrade version of JPEG:
1. high compression rate
JPEG 2000 compression performances have improved about 20% than JPEG, that is to say that under same compression ratio, the distortion of 2000 pairs of images of JPEG is littler by about 20% than JPEG.Simultaneously, use the stability of a system of JPEG 2000 good, operate steadily, anti-interference is good, easy operating.
2. support simultaneously to diminish and lossless compress
JPEG only supports lossy compression method, and JPEG 2000 can support lossless compress.In actual applications, all be very suitable for adopting JPEG 2000 compressions such as important images such as satellite remote sensing images, medical image, historical relic photos.
3. realized progressive transmission (progressive transmission)
This is 2000 extremely important features of JPEG, and the profile that it can first images progressively transmits data then, improve constantly picture quality, allow image by dim to clear demonstration, and needn't be as present JPEG, show slowly that from top to bottom this is of great importance in Network Transmission.
4. support " area-of-interest " (Region Of Interest (ROI))
The user is the compression quality of area-of-interest on the specify image arbitrarily, can also select the part of appointment to decompress earlier, gives prominence to the key points thereby make.The advantage of this method is that it combines the subjective demand of recipient to compression, has realized interactive compression.
JPEG 2000 and traditional JPEG maximum different are that it has abandoned the block coding mode based on discrete cosine transform that JPEG adopted, then adopt the multiresolution coded system based on wavelet transformation.
Wavelet transformation is the modern spectrum analysis instrument, and it can investigate the frequency domain character of local time domain process, can investigate the temporal signatures of local frequency domain process again, even therefore for non-stationary process, it is also handy to deal with.It can be a series of wavelet coefficients with image transform, and these coefficients can efficiently be compressed and store, and in addition, the rough edge of small echo is represent images better, because it has eliminated the blocking artifact that the DCT compression generally has.
The locality of small echo on space and frequency domain is the locality on the statistical significance.Here the locality of saying, it is local referring to an actual image space scope that involves of conversion coefficient.Thereby, recover certain part in the image fully, do not need all codings all accurately to be kept, only need there be error just passable corresponding to its part coding.So, can realize the area-of-interest compression.JPEG 2000 supports the shaping small echo simultaneously, thereby can realize lossless compress.
Small echo is based upon on three main basic theories, is respectively hierarchy type coding (pyramidcoding), bank of filters theory (filter bank theory) and sub-band coding (subbandcoding), and wavelet transformation combines this three technology.The signal that wavelet transformation can be formed the different frequency of various weave ins resolves into the signal of frequency inequality, so can effectively be applied to coding, decoding, detection edge, packed data, and with the nonlinear problem linearisation.The signal of the time-domain and frequency-domain of good analysis part remedies the disappearance in the Fu Liye conversion.
The following describes the principle that strengthens video image based on JPEG 2000 code streams of the present invention.
Motion JPEG 2000 is a kind of new international standards that are used for encoding moving pictures.Use the intraframe coding technology identical with JPEG 2000.Code stream is packed in advance by frame by frame.When network congestion occurring, router abandons some unessential part selectively, and only transmits the pith that is used for image reconstruction.So-called unessential packet is meant people's visual impact less, the higher or packet more clearly of image resolution ratio.Losing of high-frequency information will be alleviated the burden of Network Transmission, but can reduce the quality of image to a certain extent.
In view of Motion JPEG 2000 adopts the intraframe coding technology, the frame-to-frame correlation between each frame of the moving image behind the coding still exists, and the information that a certain two field picture is lost in transmission course can't have influence on other frame.Therefore, can recover the data of losing at the associated frame that the decoding end utilization is not subjected to view data loss to a certain extent, thereby strengthen picture quality.
According to embodiments of the invention, for example, suppose that the view data loss has appearred in present frame, and reference frame remains intact relatively receiving in the process of image by the Internet.At first between present frame and reference frame, make motion estimation.Suppose that each frame data that receives obtain frame sequence I after decoding
nAccording to the progressive transmission principle of JPEG 2000, the data volume that receives is many more, and the image restored quality is just good more.Therefore, if the data volume that detects a certain frame at receiving terminal is significantly less than the data volume of other frame, then in this frame loss of data may appear.In this case, need carry out the figure image intensifying to the present frame that receives.
Video image is carried out the enhancing process of wavelet transform below in conjunction with Fig. 2 explanation.
Process need to the figure image intensifying utilizes reference frame, and this reference frame should be a certain frame that is not subjected to data degradation that has received.In order to carry out the figure image intensifying, at receiving terminal buffer need be set and store and be used for reference frame that the picture frame that data are lost is strengthened, that is, storage one frame is not subjected to the picture signal of data degradation in buffer.Judge in receiving course whether the new picture frame that receives is subjected to data degradation.If a new frame is not subjected to data degradation, an available new frame refresh buffer.
For the ease of discussing, suppose the present frame of reception in the example below, for example, I
nView data incur loss.Should be appreciated that view data is not limited to present frame by intercepted frame in network transmission process.Can utilize, for example, the former frame I that is not subjected to data degradation of buffer storage
N-1Frame is as the reference frame.Use x
n(i, j) the spatial domain signal of the n frame of loss of data appears in expression, and through after the multistage DWT conversion, the wavelet coefficient in the subband b is expressed as y
b n(i, j), b ∈ kLL, and kHL, kLH, kHH}, wherein kHL (k is a positive integer) is illustrated in the k layer and the row signal is carried out high-pass filtering in decomposing, and column signal is carried out the result of interval sampling after the low-pass filtering.
For the purpose of narrating conveniently, suppose that every two field picture is all only through one-level wavelet transform (DWT).Because transmission is moving image, for the picture frame that is subjected to data degradation is carried out the figure image intensifying, at first needs to find the matched data piece of reference frame and present frame.As shown in Figure 2, in the LL subband with I
nFrame frequency territory signal is in the data block that is divided into the individual pixel of the M*N of fixed size (M and N are positive integers), and data block ground is at I one by one
N-1The LL subband and it is through carrying out motion search and coupling (concrete grammar will after detailed description) in three LL subbands after motion excursion.If obtain coupling, then use I
N-1This data block of frame is to I
nThe data of frame strengthen.If do not obtain coupling, then abandon this data block of reference frame.
Because the motility of image, exist certain skew between the relevant data block between reference frame and the present frame that incurs loss.In this case, carry out the figure image intensifying, need the pixel of motion picture frame for the data block that finds coupling.In the present embodiment, the process of search-related data piece is to be undertaken by the data block of mobile reference frame.
In order to search for corresponding data block, to I
N-1The frequency-region signal of frame carries out inverse discrete wavelet transform (IDWT), so that obtain I
N-1The spatial domain signal x of frame
N-1(i, j).Next, with spatial domain signal x
N-1(i, j) along the x axle, each moves a pixel y axle and x=y direction, obtains three width of cloth image x
N-1(i-1, j), x
N-1(i, j-1) and x
N-1(i-1, j-1).Respectively this three width of cloth image is carried out the DWT conversion, obtain three I through displacement
N-1The frequency-region signal of frame.With the signal indication after the conversion is y
B (1) N-1(i, j), y
B (2) N-1(i, j) and y
B (3) N-1(i, j), and
Why carry out so mobile be that if a data block mobile odd number of pixels between two frames, the then down-sampled sample that remains between two frames afterwards is supplementary set each other because image will carry out down-sampledly the DWT conversion after.Each moves a pixel in x, y and x=y direction, adds the odd even combination that image self has just comprised data block all motions on the both direction on plane.
With the frequency-region signal y in the LL subband of the present frame that receives
LL n(i j) is divided into the data block of fixed size, for example 8*8.Then one by one at four LL subband y of reference frame
LL (0) N-1(i, j), y
LL (1) N-1(i, j), y
LL (2) N-1(i, j), y
LL (3) N-1(i, j) the middle search data block the most similar to this each data block of present frame.
Search procedure can adopt three-step approach, judges that two data blocks are whether similar can adopt maximum absolute distance (MAD) rule, promptly calculates the absolute value sum of the difference between the frequency-region signal of corresponding subband of reference frame of present frame and displacement, that is, calculate
Find when making this get minimum value (di, dj, k).This minimum value and pre-set threshold are compared.If this minimum value, then shows two data blocks greater than preset threshold and does not match.If this minimum value is less than or equal to preset threshold, then should (di, dj k) note as motion vector.This motion compensation process is called as low strap displacement method (Low-Band-Shift), is characterized in more still obtaining accurate motion vectors in the information dropout of high-frequency sub-band.
In reference frame, found certain wavelet coefficient y with present frame
LL n(i, j) Dui Ying motion vector (di, dj, k) after, its wavelet coefficient corresponding in reference frame then is y
LL (k) N-1(i+di, j+dj).After this, can strengthen lost data in the present frame.As an example, for example, if in transmission course y
LL n(i j) has lost p least significant bit of present frame, receiving terminal actual receive then be
Can obtain the y without loss of following formula (1) expression this moment
LL n(i, distribution j):
Be used in the listed scope of following formula and y
LL (k N-1(i+di, j+dj) immediate value is as the wavelet coefficient to the present frame that incurs loss
The enhancing result.Its concrete operations are as follows:
If the wavelet coefficient y in the reference frame
LL (k N-1(i+di, j+dj) in except that a last p least significant bit all with
Identical, that is:
Then can make
As
The enhancing result.That is to say, use y
LL (k N-1(i+di, last p least significant bit j+dj) compensate the present frame that incurs loss
If the wavelet coefficient y in the reference frame
LL (k N-1(i+di, j+dj) in except that a last p least significant bit still with
Difference is arranged, then two end points in the interval determined of the inequality in the formula (1) relatively.In these two end points, select apart from y
LL (k N-1(i+di, j+dj) nearer point is as the enhancing result to present frame.
In the worst case, the y of present frame
LL n(i j) has lost its all significance bit (for example, the data in the whole subband are all lost), in this case, then directly makes in the present frame
According to the character of wavelet transform,, do not need again in former frame, to take exercises estimation and to search corresponding data block for three high-frequency sub-band HL, LH and the HH of image.If certain wavelet coefficient y
LL n(i, j) (di, dj k), then for all subband b, then have y to respective motion vectors
b n(i, j) Dui Ying motion vector all be (di, dj, k).In this case, can use the coefficient y in the reference frame
B (k) N-1(i+di is j+dj) as the coefficient y that strengthens the present frame of obliterated data with reference to amount
b n(i, j).
Carry out after the enhancement process through wavelet coefficient the present frame of obliterated data, wavelet coefficient after strengthening is carried out inverse discrete wavelet transform and restores such as direct current, the reprocessing of color space inverse transformation etc. and so on just can recover video image through the reconstructed image after strengthening at receiving terminal.
Be that example has illustrated that the frame-to-frame correlation of utilizing image compensates the method for enhancing to video image with one-level DWT for simplicity above.Generally speaking, a two field picture will be through the DWT conversion of 3-5 level before coding.Above-mentioned algorithm can be generalized to the DWT conversion of m level.Its process is as follows:
1. according to the method for one-level DWT conversion, to I
nLL
m, HL
m, LH
m, HH
m(m is an integer) subband strengthens.
2. four sons after strengthening are brought into the contrary DWT conversion of row, to obtain m level DWT conversion low frequency sub-band LL before
M-1
3.LL
M-1Be a subband after the enhancing, on this basis can be to other three subband HL of m-1 level
M-1, LH
M-1, HH
M-1Compensate enhancing.Its Enhancement Method and the similar process that illustrates previously.In addition, be to be noted that can with the motion vector of upper level (di, dj, k)
mAccelerate search to motion vector at the corresponding levels.
4. aforesaid thus mode strengthens the picture frame that is subjected to loss of data step by step according to resolution order from low to high, finishes up to the entire process process.
Fig. 3 shows employing image enchancing method according to the present invention carries out the image after the enhancement process to image Y-PSNR gain.This example is tested Moving Picture Experts Group-2 test pattern " moving object and calendar ".The parameter that is adopted is as follows:
Video sequence: moving object and calendar, choose 300 two field pictures, the 352*288 pixel/frame, image adopts progressive scan mode.
Wavelet basis: the 9/7 full mold bank of filters that adopts JPEG 2000 Part 1 to provide; Low-frequency band displacement (Low-Band-Shift) is adopted in motion compensation.
Compression ratio is 2.0bpp.Wherein odd-numbered frame has been carried out truncation, the code stream after blocking is equivalent to 1.0bpp.After adopting method of the present invention that the odd-numbered frame of blocking is carried out enhancement process, can obtain the Y-PSNR of average out to 3.12dB.Abscissa among Fig. 3 is represented frame number, and ordinate is represented Y-PSNR.Y-PSNR after the curve representation process figure image intensifying that blank square frame among the figure is linked to be, the Y-PSNR of the curve representation original image that black box is linked to be.Can find out obviously that from Fig. 3 the picture quality before handling through the image ratio of enhancement process is by obvious improvement.
So far described the present invention in conjunction with specific embodiment, be noted that, below not breaking away from, can change and change the present invention under the situation of the spirit or scope of claim in this description of doing as an illustration purpose only.
Claims (8)
1. video image enhancing method comprises step:
Foundation is used for the video frame image that data are lost is strengthened the image reference frame of compensation;
The loss of data frame is divided into size is the data block of M * N pixel, wherein M and N are positive integer;
Utilize inverse discrete wavelet transform to convert the frequency-region signal of reference frame to the spatial domain signal;
At the x axle, the direction of y axle and x=y moves a pixel with resulting described spatial domain signal, and the picture signal of resulting three moving directions is carried out wavelet transform to obtain the frequency-region signal of reference frame through displacement;
In the estimation of between reference frame and the loss of data frame of displacement, taking exercises, so that in reference frame, occur the data block that the data block of loss of data is mated in search and the loss of data frame;
Be used in the matched data piece that searches in the reference frame and strengthen the data block of losing in the loss of data frame.
2. video image enhancing method according to claim 1 is characterized in that further being included in and sets up after the described image reference frame, with the step of described reference frame storing in memory.
3. video image enhancing method according to claim 1 and 2 is characterized in that described step of taking exercises estimation between reference frame and loss of data frame is that the frequency-region signal of reference frame and the frequency-region signal of loss of data frame are carried out motion search and coupling.
4. video image enhancing method according to claim 1 is characterized in that at the video image transmitting terminal carrying out intraframe coding through the moving image frame of one-level wavelet transform at least.
5. video image enhancing method according to claim 1, it is characterized in that further comprising the absolute value that calculates through the difference between the frequency-region signal of the frequency-region signal of the mobile image of the reference frame of wavelet transform and loss of data frame, and with the minimum value that obtains and the step of pre-set threshold comparison.
6. video image enhancing method according to claim 5 is characterized in that not matching if described minimum value, then shows two data blocks of reference frame and loss of data frame greater than preset threshold; If this minimum value is less than or equal to preset threshold, then writes down the motion vector of described reference frame displacement and strengthen the data block of described loss of data frame with wavelet coefficient corresponding in the described reference frame with this motion vector.
7. video image enhancing method according to claim 6, it is characterized in that if having only p least significant bit in the wavelet coefficient in the reference frame different with the least significant bit of the wavelet coefficient of loss of data frame, then strengthen the least significant bit of the wavelet coefficient of loss of data frame with the least significant bit in the wavelet coefficient of reference frame, otherwise in the dynamic range that allows with the wavelet coefficient of loss of data frame and the immediate numerical value of reference frame strengthen the wavelet coefficient of described loss of data frame.
8. according to claim 1,2,4, the described video image enhancing method of in 5 to 7 any one, it is characterized in that further being included in the wavelet coefficient after the enhancing of the described loss of data frame of obliterated data is carried out inverse discrete wavelet transform and direct current restores, the color space inverse transformation is to recover the step of video image through the reconstructed image after strengthening.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 03149199 CN100477789C (en) | 2003-06-20 | 2003-06-20 | A method of inter-frame reinforcement based on Moving Joint Picture Experts Group 2000 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 03149199 CN100477789C (en) | 2003-06-20 | 2003-06-20 | A method of inter-frame reinforcement based on Moving Joint Picture Experts Group 2000 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1568011A CN1568011A (en) | 2005-01-19 |
| CN100477789C true CN100477789C (en) | 2009-04-08 |
Family
ID=34472486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 03149199 Expired - Fee Related CN100477789C (en) | 2003-06-20 | 2003-06-20 | A method of inter-frame reinforcement based on Moving Joint Picture Experts Group 2000 |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100477789C (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8072950B2 (en) | 2006-07-05 | 2011-12-06 | Samsung Electronics Co., Ltd. | Collaborative mobile ad hoc network infrastructure |
| KR20130090881A (en) | 2010-07-30 | 2013-08-14 | 톰슨 라이센싱 | Method and apparatus for measuring video quality |
| CN108596855A (en) * | 2018-04-28 | 2018-09-28 | 国信优易数据有限公司 | A kind of video image quality Enhancement Method, device and video picture quality enhancement method |
| WO2023133888A1 (en) * | 2022-01-17 | 2023-07-20 | 深圳市大疆创新科技有限公司 | Image processing method and apparatus, remote control device, system, and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4989088A (en) * | 1988-12-15 | 1991-01-29 | Kokusai Denshin Denwa Co., Ltd. | Error correction system used in picture transmission |
-
2003
- 2003-06-20 CN CN 03149199 patent/CN100477789C/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4989088A (en) * | 1988-12-15 | 1991-01-29 | Kokusai Denshin Denwa Co., Ltd. | Error correction system used in picture transmission |
Non-Patent Citations (2)
| Title |
|---|
| 基于小波变换的块匹配运动估计方法. 龚涛，丁润涛.天津大学学报，第36卷第3期. 2003 |
| 基于小波变换的块匹配运动估计方法. 龚涛，丁润涛.天津大学学报，第36卷第3期. 2003 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1568011A (en) | 2005-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1926876B (en) | Method for coding and decoding an image sequence encoded with spatial and temporal scalability | |
| US20080152011A1 (en) | Wavelet based multiresolution video representation with spatially scalable motion vectors | |
| Sathiyanathan | Medical image compression using view compensated wavelet transform | |
| KR100678909B1 (en) | Video coding method and apparatus for reducing mismatch between encoder and decoder | |
| JP2006521039A (en) | 3D wavelet video coding using motion-compensated temporal filtering in overcomplete wavelet expansion | |
| KR20000059799A (en) | Device and method for motion compensation coding using wavelet coding | |
| Surabhi et al. | Image compression techniques: a review | |
| CN100409693C (en) | Orthogonal transformation method for image and video compression | |
| Kalavathi et al. | A wavelet based image compression with RLC encoder | |
| Bhammar et al. | Survey of various image compression techniques | |
| JP2001523928A (en) | Video information compression apparatus and method | |
| US8331708B2 (en) | Method and apparatus for a multidimensional discrete multiwavelet transform | |
| Kabir et al. | Edge-based transformation and entropy coding for lossless image compression | |
| CN100534182C (en) | Method for coding a video image taking into account the part relating to a component of a movement vector | |
| CN100477789C (en) | A method of inter-frame reinforcement based on Moving Joint Picture Experts Group 2000 | |
| CN1825894A (en) | All phase cosine double orthogonal transformation and JPEG improving method | |
| Wu et al. | Comparisons of Threshold EZW and SPIHT Wavelets Based Image Compression Methods | |
| CN104079930A (en) | Achieving method of remote-sensing image compression system | |
| CN1689045A (en) | L-frames with both filtered and unfilterd regions for motion comensated temporal filtering in wavelet based coding | |
| WO2007027012A1 (en) | Video coding method and apparatus for reducing mismatch between encoder and decoder | |
| KR100361804B1 (en) | Apparatus and method for compressing and depressing movable images using Wavelet theory | |
| Ranjeeta et al. | Image compression: an overview | |
| Phakade et al. | Video compression using hybrid DCT-DWT algorithm | |
| Hassen et al. | The 5/3 and 9/7 wavelet filters study in a sub-bands image coding | |
| Pavithra et al. | An image representation scheme by hybrid compressive sensing |
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 | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090408 Termination date: 20120620 |