CN109561303A - A kind of prediction technique based on video compress - Google Patents
A kind of prediction technique based on video compress Download PDFInfo
- Publication number
- CN109561303A CN109561303A CN201811260616.7A CN201811260616A CN109561303A CN 109561303 A CN109561303 A CN 109561303A CN 201811260616 A CN201811260616 A CN 201811260616A CN 109561303 A CN109561303 A CN 109561303A
- Authority
- CN
- China
- Prior art keywords
- predicted
- value
- residual
- pixel
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- H04N19/103—Selection of coding mode or of prediction 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
- H04N19/169—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
- H04N19/17—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 being an image region, e.g. an object
- H04N19/176—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 being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- 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
- H04N19/182—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 being a pixel
-
- 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
- H04N19/184—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 being bits, e.g. of the compressed video stream
-
- 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
- H04N19/50—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using 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
- H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
- H04N19/96—Tree coding, e.g. quad-tree coding
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The present invention relates to a kind of prediction techniques based on video compress, comprising: obtains the first standard deviation and the second standard deviation of MB to be predicted；Using the first prediction residual of MB to be predicted obtain MB to be predicted the first residual absolute value and；Using the second prediction residual of MB to be predicted obtain MB to be predicted the second residual absolute value and；According to the first standard deviation and the first residual absolute value and the first residual error subjectivity of acquisition and；According to the second standard deviation and the second residual absolute value and the second residual error subjectivity of acquisition and；According to subjective and with the second residual error subjectivity and determining MB to be predicted final prediction technique of the first residual error.The present invention chooses a variety of prediction techniques for a logical image to be compressed, image to be compressed selects a kind of prediction technique to be predicted according to the correlation adaptivity of texture, so that image to be compressed targetedly can select prediction technique to be predicted, the quality of video compress is greatly improved.
Description
Technical field
The present invention relates to a kind of technical field of video compression, in particular to a kind of prediction technique based on video compress.
Background technique
Video processing technique be along with video from analog to digitalization transformation during obtain it is booming.With
People to the clarity of video image, fluency, the requirement spent is more and more harsher in real time, become one it is very powerful and exceedingly arrogant
Technology.Simultaneously because the prosperity of the industries such as current periphery industry such as Internet, display equipment, has also pushed video to handle skill
The development of art.Video processing technique can be refined as, image enhancement technology, video compress decompression technique, digital video broadcasting
The subdisciplines such as technology are applied to the various industries such as communication, family and personal entertainment, security protection, medical treatment, military affairs.Wherein video pressure
Contracting technology is especially had an optimistic view of by professional person.
So far, Video coding standard is numerous, there is static JPEG, JPEG2000, have M-JPEG, MPEG1 of movement, MPEG2,
MPEG4, H.261, H.263, WMV-HD, VC1 etc. and recently awfully hot door H.264 and the AVS of domestic independent intellectual property right etc.
Deng.Since H.264 high compression digital video decoding standard, direction prediction has been encoded into block coding in video frame in frame
Mainstream technology.
But when the texture of the image to be compressed in video is more complicated, existing prediction technique is limited to limited
Information adequately cannot effectively predict image to be compressed using the correlation between texture, to influence video compress
Quality.
Summary of the invention
It therefore, is solve technological deficiency of the existing technology and deficiency, the present invention proposes a kind of based on video compress
Prediction technique.
Specifically, a kind of prediction technique based on video compress that one embodiment of the invention proposes, comprising:
Obtain the first standard deviation and the second standard deviation of MB to be predicted；
Using the first prediction residual of the MB to be predicted obtain the MB to be predicted the first residual absolute value and；
Using the second prediction residual of the MB to be predicted obtain the MB to be predicted the second residual absolute value and；
According to first standard deviation and first residual absolute value and the first residual error subjectivity of acquisition and；
According to second standard deviation and first residual absolute value and the second residual error subjectivity of acquisition and；
According to subjective and with the second residual error subjectivity and the determining MB to be predicted final prediction of first residual error
Method.
In one embodiment of the invention, the first standard deviation and the second standard deviation of the MB to be predicted are obtained, comprising:
Obtain the first prediction residual and the first mean residual of the MB to be predicted；
First standard deviation is determined by first prediction residual and first mean residual；
Obtain the second prediction residual and the second mean residual of the MB to be predicted；
Second standard deviation is determined by second prediction residual and second mean residual.
In one embodiment of the invention, the MB to be predicted is obtained using the first prediction residual of the MB to be predicted
The first residual absolute value and, comprising:
Determine that the current pixel of the MB to be predicted has K pixel component, wherein K is the natural number greater than zero；
Obtain the first weighted gradient value of the pixel component；
The second weighted gradient value of the pixel component is obtained by the first weighted gradient value；
The reference pixel value of the pixel component is determined by the second weighted gradient value；
Difference is asked to obtain first prediction residual pixel value of the pixel component and the reference pixel value；
According to first prediction residual obtain the MB to be predicted the first residual absolute value and.
In one embodiment of the invention, the first weighted gradient value of the pixel component is obtained, comprising:
By surrounding's component of the pixel component, N number of grain direction gradient value of the pixel component is determined, wherein N
For the natural number greater than zero；
It is weighted processing using grain direction gradient value of first weighting coefficient to each pixel component, obtains institute
State the first weighted gradient value of each pixel component.
In one embodiment of the invention, obtain the pixel component by the first weighted gradient value second adds
Weigh gradient value, comprising:
Choose the minimum value of the first weighted gradient value；
Place is weighted using minimum value of second weighting coefficient to the first weighted gradient value of each pixel component
Reason obtains the second weighted gradient value of each pixel component.
In one embodiment of the invention, the reference image of the pixel component is determined by the second weighted gradient value
Element value, comprising:
The reference direction of the pixel component is determined by the minimum value of the second weighted gradient value；
It is weighted processing using component pixel of the third weighting coefficient to the reference direction, obtains the pixel point
The reference pixel value of amount.
In one embodiment of the invention, the MB to be predicted is obtained using the second prediction residual of the MB to be predicted
The second residual absolute value and, comprising:
The MB to be predicted is divided into multiple sub- MB to be predicted using quadtree approach；
The prediction residual of the MB to be predicted and the prediction residual of the first bit number and the sub- MB to be predicted are obtained respectively
With the second bit number；
Pass through the prediction residual of the prediction residual of the MB to be predicted and the first bit number and the sub- MB to be predicted and the
Two bit numbers determine the second prediction residual of the MB to be predicted；
Judge whether the sub- MB to be predicted continues to divide, if so, it is described to pre- to continue segmentation according to QuadTree algorithm
Sub- MB is surveyed, if it is not, then terminating the segmentation of the sub- MB to be predicted, institute is obtained by the second prediction residual of the sub- MB to be predicted
State MB to be predicted the second residual absolute value and.
In one embodiment of the invention, prediction residual and the first bit number and the institute of the MB to be predicted are obtained respectively
State the prediction residual and the second bit number of sub- MB to be predicted, comprising:
All pixels component value in the MB to be predicted is individually subtracted to the minimum of pixel component value in the MB to be predicted
Value, obtains the corresponding prediction residual of all pixels component in the MB to be predicted；
Using the first least number of bits of the MB to be predicted, the data bit depth of the MB to be predicted with described to pre-
The quantity for surveying the pixel component of MB determines first bit number；
Pixel component value in the sub- MB to be predicted is individually subtracted in all pixels component value in the sub- MB to be predicted
Minimum value obtains the corresponding prediction residual of all pixels component in the sub- MB to be predicted；
Using the second least number of bits of the sub- MB to be predicted, the data bit depth of the sub- MB to be predicted with it is described
The quantity of the pixel component of sub- MB to be predicted determines second bit number.
In one embodiment of the invention, by the prediction residual of the MB to be predicted and the first bit number and it is described to
Predict that the prediction residual of sub- MB and the second bit number determine the second prediction residual of the MB to be predicted, comprising:
The first weighted value of the MB to be predicted is determined according to the prediction residual of the MB to be predicted and the first bit number；
The second weighted value of the MB to be predicted is determined according to the prediction residual of the sub- MB to be predicted and the second bit number；
First weighted value and second weighted value are judged；
If first weighted value is greater than second weighted value, continue to divide the sub- MB to be predicted；
If first weighted value is less than second weighted value, the prediction residual conduct of the sub- MB to be predicted is chosen
The second prediction residual of the MB to be predicted.
In one embodiment of the invention, subjective and determining according to the first residual error subjectivity and with second residual error
The final prediction technique of the MB to be predicted, comprising:
Choose first residual error it is subjective and with the second residual error subjectivity and in minimum value, it is true by the minimum value
The final prediction technique of the fixed MB to be predicted.
Based on this, the present invention has following advantage:
The present invention has chosen a variety of prediction techniques for a logical image to be compressed and predicts, image to be compressed can basis
The selection one of which prediction technique of the correlation adaptivity of texture is predicted, so that image to be compressed can be directed to
Property selection prediction technique predicted, greatly improve the quality of video compress, and by selecting optimal prediction side
Method, which carries out prediction, can further decrease theoretical limit entropy.
Through the following detailed description with reference to the accompanying drawings, other aspects of the invention and feature become obvious.But it should know
Road, which is only the purpose design explained, not as the restriction of the scope of the present invention, this is because it should refer to
Appended claims.It should also be noted that unless otherwise noted, it is not necessary to which scale attached drawing, they only try hard to concept
Ground illustrates structure and process described herein.
Detailed description of the invention
Below in conjunction with attached drawing, specific embodiments of the present invention will be described in detail.
Fig. 1 is a kind of prediction technique flow diagram based on video compress provided in an embodiment of the present invention；
Fig. 2 is a kind of algorithm of the adaptive direction prediction technique of Pixel-level multi -components reference provided in an embodiment of the present invention
Schematic illustration；
Fig. 3 is a kind of reference pixel position view provided in an embodiment of the present invention；
Fig. 4 is that a kind of gradient value provided in an embodiment of the present invention calculates schematic diagram；
Fig. 5 is the calculation of the adaptive direction prediction technique of another Pixel-level multi -components reference provided in an embodiment of the present invention
Method schematic illustration；
Fig. 6 is a kind of schematic diagram of the dividing method based on quaternary tree provided in an embodiment of the present invention；
Fig. 7 is a kind of schematic diagram of macroblock partition mode to be predicted provided in an embodiment of the present invention；
Fig. 8 is the schematic diagram of another kind provided in an embodiment of the present invention macroblock partition mode to be predicted.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing to the present invention
Specific embodiment be described in detail.
Embodiment one
Referring to Figure 1, Fig. 1 is a kind of prediction technique process signal based on video compress provided in an embodiment of the present invention
Figure.The prediction technique includes the following steps:
Step 1, the first standard deviation and the second standard deviation for obtaining MB to be predicted；
Step 2, the first residual absolute value that the MB to be predicted is obtained using the first prediction residual of the MB to be predicted
With；
Step 3, the second residual absolute value that the MB to be predicted is obtained using the second prediction residual of the MB to be predicted
With；
Step 4, according to first standard deviation and first residual absolute value and obtain the first residual error it is subjective and；
Step 5, according to second standard deviation and second residual absolute value and obtain the second residual error it is subjective and；
Step 6, according to first residual error it is subjective and with second residual error it is subjective and determine the MB to be predicted most
Whole prediction technique.
Specifically, in order to better illustrate the prediction technique based on video compress, the present embodiment will to the prediction technique into
Row detailed description:
S1, the size of MB to be predicted is set as m*n, wherein m and n is the natural number greater than zero.
S2, MB to be predicted is predicted according to the adaptive direction prediction technique that Pixel-level multi -components refer to, and according to
Residual absolute value and calculation formula calculate the first residual absolute value and SAD of MB to be predicted1, the first residual absolute value and SAD1It asks
Be the first prediction residual all in MB to be predicted absolute value and, wherein residual absolute value and calculation formula are as follows:
Wherein, SAD be residual absolute value and, ABS is to seek absolute value, and Res is prediction residual, and k is corresponding to prediction technique
Serial number, Pixel-level multi -components reference the corresponding k of adaptive direction prediction technique be 1, calculate the first residual absolute value SAD1
When corresponding all first prediction residuals for MB to be predicted.
S3, MB to be predicted is predicted according to the video compress prediction technique based on quaternary tree, and absolute according to residual error
Value and calculation formula calculate the second residual absolute value and SAD of MB to be predicted2, the first residual absolute value SAD2Acquire is to pre-
Survey MB all second prediction residuals absolute value and, the corresponding k of video compress prediction technique based on quaternary tree be 2.
S4, the adaptive direction prediction technique corresponding that the reference of Pixel-level multi -components is calculated using standard deviation calculation formula
One standard deviation E1, the corresponding second standard deviation E of video compress prediction technique based on quaternary tree2, wherein standard deviation calculation formula
Are as follows:
Wherein, E is standard deviation, and ABS is to seek absolute value, and Res is prediction residual, and AVE is mean residual, and k is prediction technique
Corresponding serial number, the value of k are 1 and 2, when k is 1, the corresponding adaptive direction prediction for the reference of Pixel-level multi -components
Method, when k is 2, corresponding is the video compress prediction technique based on quaternary tree.Calculate the first standard deviation E1When corresponding be
First prediction residual calculates the second standard deviation E2When it is corresponding be the second prediction residual,.
Step 5, finally according to SAD1、SAD2、E1And E2The case where, divide scene configuration weight coefficient a3And a4, according to SUBD
(subjectivedifference, residual error subjectivity and) calculation formula calculate SUBD, wherein Pixel-level multi -components reference it is adaptive
Direction prediction method is answered to correspond to SUBD1, the video compress prediction technique based on quaternary tree corresponds to SUBD2。
SUBDk=a1×SADk+a2×Ek
Wherein, a3With a4 be weight coefficient, SAD be residual absolute value and, E is standard deviation, and Res is prediction residual, and k is pre-
Serial number corresponding to survey mode, wherein the adaptive direction prediction technique of respective pixel grade multi -components reference when k is 1, when k is 2
The corresponding video compress prediction technique based on quaternary tree.
Compare SUBD1And SUBD2, selecting prediction technique corresponding to wherein the smallest SUBD is final prediction technique, is used
The prediction residual of the final prediction technique is final prediction residual；And the prediction residual of the final prediction technique is transmitted in code stream
With additional mark position, additional mark position is serial number corresponding to the final prediction technique, respective pixel grade when additional mark position is 1
The adaptive direction prediction technique of multi -components reference, the corresponding video compress prediction based on quaternary tree when being 2 that additional mark position is k
Method.
When the texture of image to be compressed is complex, it is predicted respectively by two kinds of prediction techniques, according to pre-
It surveys result and chooses optimal prediction technique from two kinds of prediction techniques as final prediction technique, theory can be further decreased
Limit entropy carries out the selection of prediction technique according to the textural characteristics of image to be compressed, improves the adaptivity of prediction technique, and
And the compression quality of the image to be compressed of texture complexity is improved, improve prediction effect.
The present invention has chosen two kinds of prediction techniques for a logical image to be compressed and predicts that image to be compressed can basis
The selection one of which prediction technique of the correlation adaptivity of texture is predicted, so that image to be compressed can be directed to
Property selection prediction technique predicted, greatly improve the quality of video compress, and by selecting optimal prediction side
Method, which carries out prediction, can further decrease theoretical limit entropy.
Embodiment two
Fig. 2 and Fig. 3 are referred to, Fig. 2 is a kind of adaptive side of Pixel-level multi -components reference provided in an embodiment of the present invention
To the algorithm principle schematic diagram of prediction technique, Fig. 3 is a kind of reference pixel position view provided in an embodiment of the present invention.This reality
Apply example on the basis of the above embodiments to Pixel-level multi -components proposed by the present invention reference adaptive direction prediction technique into
Row detailed description, the prediction technique include the following steps:
S1, the size for defining MB to be predicted are m*n, wherein m and n is the natural number greater than zero；
S2, the current pixel for defining MB to be predicted have K pixel component, wherein K is the natural number greater than zero, and the K is a
Pixel component is respectively pixel component 1, pixel component 2 ... pixel component K；
S3, each pixel component for current pixel are determined each by the surrounding pixel component of each pixel component
N number of grain direction gradient value G1~GN of pixel component, wherein N is the natural number greater than zero；
Preferably, surrounding's component of each pixel component, can be adjacent with the pixel component, can also be non-conterminous；Such as Fig. 3 institute
Show, Cur represents current pixel component, and surrounding pixel component can be GHIK, or ABCDEFJ.
S4, N number of grain direction gradient value G1~GN of each pixel component is weighted processing, and (G1~GN both represents line
The size for managing direction gradient value, also represents the direction of grain direction gradient value), obtain N number of weighted place of grain direction gradient value
The first weighted gradient value BG after reason, the calculation formula of the first weighted gradient value BG are as follows:
BGi=w1*G1+w2*G2+ ...+wN*GN (i=1 ... K)
Wherein, w1, w2 ... wN are the first weighting coefficient, and w1, w2 ... wN can be the same or different；Corresponding, BG1 is
First weighted gradient value of pixel component 1, BG2 are the first weighted gradient value of pixel component 2, and so on, BGK is pixel point
Measure the first weighted gradient value of K.
In one embodiment, w1, w2 ... wN can be the fixed value of preparatory sets itself.Further, also,
When configuring the relative size of w1, w2 ... wN, it may be considered that priori.For example, being learnt from previous experience, gradient value G1's
This direction may be more suitable the actual conditions that this image gives a forecast, then w1 can be configured to one be more suitable this image and do
The value (for example, w1 can be configured into very little) of the actual conditions of prediction, to increase the weight in this direction of gradient value G1.When
So, w1, w2 ... wN are also possible to adaptive, it can according to the actual conditions that early prediction is handled, w1, w2 are adjusted flexibly ...
The relative size of wN, specifically w1+w2+ ...+wN=1.
Preferably, the first weighted gradient value BG can use the absolute value representation of the margin of image element of corresponding pixel component, still
It is without being limited thereto.
Preferably, the value for choosing multiple groups w1, w2 ... wN, obtains multiple first weighted gradient values, the first weighted gradient value BG
It is minimized, the optimal value BGbst of the first weighted gradient value of available each pixel component.
S5, the optimal value BGbst of the first weighted gradient value of K pixel component is weighted to processing, available
The calculating of the optimal value of one weighted gradient value is weighted treated the second weighted gradient value BG ", the second weighted gradient value BG " is public
Formula is as follows:
BG " i=t1*BGbst1+t2*BGbst2+ ...+tK*BGbstK (i=1 ... K)
Wherein, t1, t2 ... tK are the second weighting coefficient, and t1, t2 ... tK can be the same or different；BGbst1 is pixel
The optimal value of first weighted gradient value of component 1, BGbst2 are the optimal value of the first weighted gradient value of pixel component 2, successively
Analogize, BGbstK is the optimal value of the first weighted gradient value of pixel component K, and BG " 1 is the second weighted gradient of pixel component 1
Value, BG " 2 are the second weighted gradient value of pixel component 2, and so on, BG " K is the second weighted gradient value of pixel component K,
Determine the second weighted gradient value BG " optimal value BG " bst of each pixel component.
Preferably, according to the relationship of each pixel component and the optimal value BGbst of corresponding first weighted gradient value, setting adds
Weight coefficient t1, t2 and t3 obtain the optimal value BGbst of the second weighted gradient value of each pixel component.
Preferably, the optimal value BGbst weighting coefficient values of the first weighted gradient value under current pixel component are maximum, and work as
The optimal value BGbst weighting coefficient values of the first weighted gradient value under other pixel components that preceding pixel component distance gradually increases
It is gradually reduced, the summation of weighting coefficient values is 1, specially t1+t2+t3=1.
Preferably, the second weighted gradient value BG " is minimized, the second weighted gradient value of available each pixel component
Optimal value BG " bst.
The corresponding direction optimal value BG " bst of second weighted gradient value is the reference direction Dir of the pixel component.
S6, the pixel value of available pixel components all in the reference direction of each pixel component is weighted processing,
The reference pixel value Ref of each pixel component is obtained, reference pixel value Ref calculation formula is as follows:
Refi=r1*cpt1+r2*cpt2+ ...+rN*cptN (i=1 ... K)
Wherein, r1, r2 ... rN are third weighting coefficient, and r1, r2 ... rN can be the same or different；Cpt1~cptN is
The pixel value of N number of available pixel component in the reference direction of each pixel component；Ref1 is the reference pixel of pixel component 1
Value, Ref2 are the reference pixel value of pixel component 2, and so on, RefK is the reference pixel value of pixel component K.
S7, the pixel value of current pixel component is subtracted to corresponding reference pixel value, available current pixel component
First prediction residual Dif；Formula is as follows:
Difi=Curcpti-Refi (i=1 ... K)
Wherein, Curcpt1 is the pixel value of pixel component 1, and Curcpt2 is the pixel value of pixel component 2, and so on,
CurcptK is the pixel value of pixel component K；Dif1 is the prediction residual of pixel component 1, and Dif2 is that the prediction of pixel component 2 is residual
Difference, and so on, DifK is the prediction residual of pixel component K.
Remaining component of S8, current pixel, repetition S3~S7 are pre- to get arrive the current pixel all pixels component first
Survey residual error.
Preferably, multi -components can also be needed with parallel processing with serial process, concrete foundation application specification scene.
The present embodiment is weighted processing by multiple pixel components to current pixel, can more reasonably determine and work as
The prediction direction of preceding pixel component can play better prediction direction rectifying effect especially when texture is more complicated.And
And this method, can texture prediction direction between multiple adjacent pixels of balanced pixel component, reduce the prediction of single pixel component
A possibility that erroneous judgement, the theoretical limit entropy of prediction is finally further decreased, the present invention can also locate parallel multiple pixel components
Reason, more conducively the parallelization processing of realization prediction technique.Relative to the time long low efficiency of serial component processing, parallel processing can
To significantly improve processing speed, conducive to the hardware realization of prediction algorithm.
Embodiment three
Fig. 4 and Fig. 5 are referred to, Fig. 4 is that a kind of gradient value provided in an embodiment of the present invention calculates schematic diagram；Fig. 5 is this hair
The algorithm principle schematic diagram of the adaptive direction prediction technique for another Pixel-level multi -components reference that bright embodiment provides.This reality
Example is applied on the basis of the above embodiments to lift the adaptive direction prediction technique of Pixel-level multi -components proposed by the present invention reference
Example description.The pixel of current pixel is divided into tri- components of Y, U, V by the present embodiment, the specific steps are as follows:
S1, the size for defining MB to be predicted are m*n, wherein m and n is the natural number greater than zero；
S2, define the current pixel of MB to be predicted there are three pixel component, respectively pixel component Y, pixel component U, as
Prime component V；
S3, each pixel is determined by surrounding's component of each pixel component for three pixel components of current pixel
3 grain direction gradient values G1, G2 and G3 of component；
Preferably for pixel component Y, pixel component U, pixel component V, respectively according to Fig.4, available ABS
It (K-H) is 45 degree of grain direction gradient values, ABS (K-G) is 90 degree of grain direction gradient values, and ABS (K-F) is 135 degree of texture sides
To gradient value, ABS (K-J) is 180 degree grain direction gradient value, wherein ABS is signed magnitude arithmetic(al).
S4,3 grain direction gradient values G1, G2, G3 of pixel component Y, pixel component U, pixel component V are added
Power processing, respectively obtains the first weighted gradient value BG of each pixel component, searches the first weighting of each pixel component
The minimum value BGmin of gradient value, the optimal value as the first weighted gradient value.
S5, the minimal gradient value of the first weighted gradient value of 3 pixel components is weighted processing, obtains the first weighting
The optimal value of gradient value is weighted treated the second weighted gradient value BG ", and find out the minimum value of the second weighted gradient value
Optimal value BG " the bst of BG " min is as the second weighted gradient value BG ".Weighted calculation is as follows:
BG " minY=0.5*BGminY+0.7*BGminU+0.7*BGminV
BG " minU=0.5*BGminY+0.3*BGminU+0.4*BGminV
BG " minY=0.5*BGminY+0.4*BGminU+0.3*BGminV
Wherein, BG " minY is the minimum value of the second weighted gradient of pixel component Y value, and BG " minU is pixel component U second
The minimum value of weighted gradient value, BG " minV are the minimum value of the second weighted gradient of pixel component V value, and BGminY is pixel component Y
The minimum value of first weighted gradient value, BGminU are the minimum value of the first weighted gradient of pixel component U value, and BGminV is pixel point
Measure the minimum value of the first weighted gradient of V value.
Wherein, the direction of BG " min is the reference direction Dir of current pixel component, i.e. DirY is the reference of pixel component Y
Direction, DirU are the reference direction of pixel component U, and DirV is the reference direction of pixel component V.
S5, by pixel component Y, pixel component U, tri- pixel components of pixel component V reference direction on two components
Pixel value is weighted, and obtains the reference pixel value Ref of tri- components of reference pixel value Ref, reference pixel value Ref calculation formula
It is as follows:
RefY=r1*cpt1+r2*cpt2
RefU=r1*cpt1+r2*cpt2
RefV=r1*cpt1+r2*cpt2
Wherein, RefY is the reference pixel value of pixel component Y, and RefU is the reference pixel value of pixel component U, and RefV is picture
The reference pixel value of prime component V, cpt1, cpt2 are the component pixel of each reference direction, and r1 and r2 are third weighting coefficient.
It preferably for any pixel component, is referred to if 45 degree, then reference pixel value REF is 0.8*I+0.2E；If
It is referred to for 90 degree, then reference pixel value is 0.8*H+0.2C；It is referred to if 135 degree, then reference pixel value is 0.8*G+
0.2A；It is referred to if 180 degree, then reference pixel value is 0.8*K+0.2J.
S36, the pixel value of current pixel component is subtracted to corresponding reference pixel value, available current pixel component
First prediction residual Res, the first prediction residual Res calculation formula are as follows:
ResY=CurcptY-RefY
ResU=CurcptU-RefU
ResV=CurcptV-RefV
Wherein, CurcptY is the pixel value of pixel component Y, and CurcptU is the pixel value of pixel component U, and CurcptV is
The pixel value of pixel component V；ResY is the first prediction residual of pixel component Y, and the first prediction that ResU is pixel component U is residual
Difference, ResV are the first prediction residual of pixel component V.
The present invention is handled by the multi-direction gradient weighting to tri- pixel components of R, G, B, can more reasonably be determined
The prediction direction of current pixel component can play better prediction direction rectifying effect especially when texture complexity.And
This method can be balanced with the texture prediction direction between tri- pixel components of position R, G, B, reduces the prediction of single pixel component
A possibility that erroneous judgement, finally further decrease the theoretical limit entropy of prediction, the present invention can also be more sharp by multi -components parallel processing
In the parallelization processing for realizing prediction technique.Relative to the time long low efficiency of serial component processing, parallel processing can be at double
Processing speed is improved, conducive to the hardware realization of prediction algorithm.
Example IV
Fig. 6~Fig. 8 is referred to, Fig. 6 is a kind of signal of the dividing method based on quaternary tree provided in an embodiment of the present invention
Figure, Fig. 7 are a kind of schematic diagram of macroblock partition mode to be predicted provided in an embodiment of the present invention, and Fig. 8 mentions for the embodiment of the present invention
The schematic diagram of the another kind macroblock partition mode to be predicted of confession.The present embodiment on the basis of the above embodiments proposes the present invention
The video compress prediction technique based on quaternary tree describe in detail.
Video usually may include a series of pictures, and each picture can be divided into or be divided into presumptive area, such as frame or
MB.When the region of video is divided into MB, according to coding method, MB or interframe MB in framing can be classified by dividing MB.Frame
Interior MB refers to the MB encoded by intra-frame predictive encoding method.Intra-frame predictive encoding method is by using wherein executing present encoding
Current image in front of be subjected to coding and decoding reconstructed blocks pixel, predict the pixel of MB to be predicted, with generate prediction
Then MB encodes the difference between the pixel of MB to be predicted and the pixel of MB to be predicted.
In the present invention, as shown in fig. 6, coded object can be the image MB of 64 × 64 specifications, or one
A 16 × 16 specification image MB, more either with the image macro of smaller or larger dimensions.For example, MB to be predicted is pressed
Recursive subdivision is carried out according to QuadTree algorithm, each MB to be predicted is divided into the sub- MB to be predicted of four same sizes.Each to
Predict whether sub- MB is further continued for being split being judged by preset algorithm.
Assuming that MB to be predicted is 64 × 64 specifications, using 64 × 64 MB to be predicted as root node, it is located at first layer.Pass through
When preset algorithm judges to need to continue segmentation, which is divided into 4 32 × 32 sub- MB to be predicted, forms the second layer.
Judge that second layer upper right sub- MB to be predicted and second layer lower-left sub- MB to be predicted need not continue to divide by preset algorithm, the
Two layers of upper left sub- MB to be predicted and second layer bottom right sub- MB needs to be predicted continue to divide, by second layer upper left sub- MB to be predicted points
4 16 × 16 sub- MB to be predicted are segmented into, second layer bottom right sub- MB to be predicted is divided into 4 16 × 16 sub- MB to be predicted,
Third layer is formed, successively recurrence, until n-th layer.As shown in fig. 7, the final segmentation side of the MB to be predicted for 64 × 64 specification
Formula.
The prediction technique of the present embodiment includes the following steps:
S1, MB to be predicted is split according to QuadTree algorithm, as shown in figure 8, to be predicted after MB to be predicted segmentation
Sub- MB is respectively the first sub- MB to be predicted, the second sub- MB to be predicted, the third sub- MB to be predicted of sub- MB and the 4th to be predicted.
S2, the first bit number and prediction residual that MB to be predicted is obtained according to original MB to be predicted, specifically, calculate to
It predicts the difference in MB in pixel component maximum value and MB to be predicted between pixel component minimum value, obtains indicating the first of difference
Least number of bits obtains the first of MB to be predicted according to the data bit depth calculation of the first least number of bits and MB to be predicted
Bit number, wherein the first bit number of MB to be predicted meets following formula:
MBIT1=M*BIT_MIN1+2*BITDETH
Wherein, MBIT1 is the first bit number of MB to be predicted, and BIT_MIN1 is the first least number of bits, and BITDEPTH is
The data bit depth of MB to be predicted, M are the pixel component quantity in MB to be predicted.
All pixels component value in MB to be predicted is individually subtracted to the minimum value of all pixels component value in MB to be predicted, is obtained
The corresponding prediction residual of all pixels component into MB to be predicted.
S3, corresponding second bit number of each sub- MB to be predicted and pre- is obtained according to sub- MB to be predicted each of after segmentation
Residual error is surveyed, specifically, calculates in the first sub- MB to be predicted in pixel component maximum value and the first sub- MB to be predicted pixel component most
The second difference between small value obtains the second least number of bits for indicating the first sub- MB to be predicted；
Calculate in the second sub- MB to be predicted in pixel component maximum value and the second sub- MB to be predicted pixel component minimum value it
Between third difference, obtain indicate the second sub- MB to be predicted third least number of bits；Calculate pixel in third sub- MB to be predicted
The 4th difference in component maximum value and third sub- MB to be predicted between pixel component minimum value obtains indicating third son to be predicted
The 4th least number of bits of MB；Calculate pixel component maximum value and pixel point in the 4th sub- MB to be predicted in the 4th sub- MB to be predicted
The 5th difference between minimum value is measured, the 5th least number of bits for indicating the 4th sub- MB to be predicted is obtained；According to the second minimum ratio
Special number, third least number of bits, the 4th least number of bits, the 5th least number of bits and MB to be predicted data bit depth gauge
Calculation obtains the second bit number, wherein the second bit number meets following formula:
MBIT2=N1*BIT_MIN2+N2*BIT_MIN3+N3*BIT_MIN4+N4*BIT_MIN5+2* BITDETH wherein,
MBIT2 is the second bit number, and BIT_MIN2 is the second least number of bits, and BIT_MIN3 is third least number of bits, BIT_MIN4
For the 4th least number of bits, BIT_MIN5 is the 5th least number of bits, and BITDEPTH is the data bit depth of MB to be predicted, N1
For pixel component quantity in the first sub- MB to be predicted, N2 is pixel component quantity in the second sub- MB to be predicted, and N3 waits for pre- for third
Pixel component quantity in sub- MB is surveyed, N4 is pixel component quantity in the 4th sub- MB to be predicted.
All pixels component in the first sub- MB to be predicted is individually subtracted in all pixels component value in first sub- MB to be predicted
The minimum value of value, all pixels component in the second sub- MB to be predicted is individually subtracted in all pixels component value in the second sub- MB to be predicted
The minimum value of value, all pixels component in third sub- MB to be predicted is individually subtracted in all pixels component value in third sub- MB to be predicted
The minimum value of value, all pixels component in the 4th sub- MB to be predicted is individually subtracted in all pixels component value in the 4th sub- MB to be predicted
The minimum value of value, corresponding second prediction residual of all pixels component in all sub- MB to be predicted after being divided.
It is S4, residual according to the prediction of the first bit number, the prediction residual of MB to be predicted, the second bit number and sub- MB to be predicted
Difference judges whether to continue to divide to MB to be predicted；If so, step 1 is jumped to, it will be each to be predicted according to recursive algorithm
A sub- MB executes step 1~step 4 respectively and is split；If it is not, then terminating the segmentation of sub- MB to be predicted.
Specifically, the first reconstructed value of MB to be predicted is obtained according to the prediction residual of MB to be predicted, ask the first reconstructed value with
The pixel value of MB to be predicted seeks absolute value of the difference, obtains the first reconstruction difference, by first rebuild difference and the first bit number into
Row weighting obtains the first weighted value of MB to be predicted, wherein the first weighted value meets following formula:
RDO1=a1*MBIT1+b1*RES1
Wherein, RDO1 is the first weighted value, and MBIT1 is the first bit number, and RES1 is the first reconstruction difference, a1And b1To add
Weight coefficient.
a1And b1Value can be preset fixed value, further, a1+b1=1, it is preferable that a1It can choose
It is 0.5, b1It can be chosen for 0.5, a1And b1Size can also be adjusted flexibly.
Wherein, reconstruction pixel component, which refers to, has compressed the pixel component that image decompression is rebuild, and rebuilds the picture of pixel component
Element value is commonly referred to as reconstructed value.Further, according to the available reconstruction pixel component of prediction residual, i.e., reference value is (each
The minimum value of macro block pixels component) add the available reconstruction pixel component of prediction residual.
The second reconstructed value of sub- MB to be predicted after being divided according to the second prediction residual asks the second reconstructed value and segmentation
The pixel value of sub- MB to be predicted afterwards seeks absolute value of the difference, obtains the second reconstruction difference, rebuilds difference and the second ratio for second
Special number is weighted the second weighted value of the sub- MB to be predicted after being divided, wherein the second weighted value meets following formula:
RDO2=a2*MBIT2+b2*RES2
Wherein, RDO2 is the second weighted value, and MBIT2 is the second bit number, and RES2 is the second reconstruction difference, a2And b2To add
Weight coefficient.
a2And b2Value can be preset fixed value, further, a2+b2=1, it is preferable that a2It can choose
It is 0.5, b2It can be chosen for 0.5, a2And b2Size can also be adjusted flexibly.
Compare the size of the first weighted value Yu the second weighted value, it, will be to pre- if the first weighted value is greater than the second weighted value
It surveys sub- MB to be split according to QuadTree algorithm, each sub- MB to be predicted is executed into step 1~step 4 respectively and is judged whether
Continue to divide, i.e., according to recursive algorithm, judges whether to third time and divide, divide for the 4th time until n-th is divided.Instead
It, MB to be predicted is without segmentation if the first weighted value is less than the second weighted value.
S5, output MB to be predicted are finally divided under level, in the prediction residual and pixel component of each sub- MB to be predicted
Minimum value, and using the prediction residual of the sub- MB to be predicted under final segmentation level as the second prediction residual.
The present invention is based on the video compress prediction technique of quaternary tree, be according to the bit number and prediction residual of MB to be predicted with
And the bit and prediction residual of sub- MB to be predicted judges whether to continue segmentation, it in this way can targetedly MB to be predicted
It is split, to can determine final segmentation according to the correlation of texture when the texture of image to be compressed is complex
Level is predicted by this partitioning scheme, can be improved compression efficiency, and improve subjective picture quality, and for multiple
Miscellaneous texture image processing, the prediction technique effect is good, treatment effeciency is high, and can reduce theoretical limit entropy.
In conclusion specific case used herein a kind of is explained to of the invention based on video compress prediction technique
It states, the above description of the embodiment is only used to help understand the method for the present invention and its core ideas；Meanwhile for this field
Those skilled in the art, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, to sum up institute
It states, the contents of this specification are not to be construed as limiting the invention, and protection scope of the present invention should be with the attached claims
It is quasi-.
Claims (10)
1. a kind of prediction technique based on video compress characterized by comprising
Obtain the first standard deviation and the second standard deviation of MB to be predicted；
Using the first prediction residual of the MB to be predicted obtain the MB to be predicted the first residual absolute value and；
Using the second prediction residual of the MB to be predicted obtain the MB to be predicted the second residual absolute value and；
According to first standard deviation and first residual absolute value and the first residual error subjectivity of acquisition and；
According to second standard deviation and second residual absolute value and the second residual error subjectivity of acquisition and；
According to subjective and with the second residual error subjectivity and the determining MB to be predicted final prediction technique of first residual error.
2. the method according to claim 1, wherein obtaining the first standard deviation and the second standard of MB to be predicted
Difference, comprising:
Obtain the first prediction residual and the first mean residual of the MB to be predicted；
First standard deviation is determined by first prediction residual and first mean residual；
Obtain the second prediction residual and the second mean residual of the MB to be predicted；
Second standard deviation is determined by second prediction residual and second mean residual.
3. the method according to claim 1, wherein the first prediction residual using the MB to be predicted obtains institute
State MB to be predicted the first residual absolute value and, comprising:
Determine that the current pixel of the MB to be predicted has K pixel component, wherein K is the natural number greater than zero；
Obtain the first weighted gradient value of the pixel component；
The second weighted gradient value of the pixel component is obtained by the first weighted gradient value；
The reference pixel value of the pixel component is determined by the second weighted gradient value；
Difference is asked to obtain first prediction residual pixel value of the pixel component and the reference pixel value；
According to first prediction residual obtain the MB to be predicted the first residual absolute value and.
4. according to the method described in claim 3, wrapping it is characterized in that, obtain the first weighted gradient value of the pixel component
It includes:
By surrounding's component of the pixel component, N number of grain direction gradient value of the pixel component is determined, wherein N is big
In zero natural number；
It is weighted processing using grain direction gradient value of first weighting coefficient to each pixel component, is obtained described every
First weighted gradient value of a pixel component.
5. according to the method described in claim 3, it is characterized in that, obtaining the pixel point by the first weighted gradient value
Second weighted gradient value of amount, comprising:
Choose the minimum value of the first weighted gradient value；
It is weighted processing using minimum value of second weighting coefficient to the first weighted gradient value of each pixel component, is obtained
Obtain the second weighted gradient value of each pixel component.
The reference direction of each pixel component is determined by the second weighted gradient value.
6. according to the method described in claim 3, it is characterized in that, determining the pixel point by the second weighted gradient value
The reference pixel value of amount, comprising:
The reference direction of the pixel component is determined by the minimum value of the second weighted gradient value；
It is weighted processing using component pixel of the third weighting coefficient to the reference direction, obtains the pixel component
Reference pixel value.
7. the method according to claim 1, wherein the second prediction residual using the MB to be predicted obtains institute
State MB to be predicted the second residual absolute value and, comprising:
The MB to be predicted is divided into multiple sub- MB to be predicted using quadtree approach；
The prediction residual of the MB to be predicted and the prediction residual and of the first bit number and the sub- MB to be predicted are obtained respectively
Two bit numbers；
Pass through the prediction residual and the second ratio of the prediction residual of the MB to be predicted and the first bit number and the sub- MB to be predicted
Special number determines the second prediction residual of the MB to be predicted；
Judge whether the sub- MB to be predicted continues to divide, if so, continuing to divide the son to be predicted according to QuadTree algorithm
MB, if it is not, then terminate the segmentation of the sub- MB to be predicted, by the second prediction residual of the sub- MB to be predicted obtain it is described to
Predict MB the second residual absolute value and.
8. the method according to the description of claim 7 is characterized in that obtaining the prediction residual and first of the MB to be predicted respectively
The prediction residual and the second bit number of bit number and the sub- MB to be predicted, comprising:
All pixels component value in the MB to be predicted is individually subtracted to the minimum value of pixel component value in the MB to be predicted, is obtained
The corresponding prediction residual of all pixels component into the MB to be predicted；
Utilize the first least number of bits of the MB to be predicted, the data bit depth of the MB to be predicted and the MB to be predicted
The quantity of pixel component determine first bit number；
All pixels component value in the sub- MB to be predicted is individually subtracted to the minimum of pixel component value in the sub- MB to be predicted
Value, obtains the corresponding prediction residual of all pixels component in the sub- MB to be predicted；
Using the second least number of bits of the sub- MB to be predicted, the data bit depth of the sub- MB to be predicted with described to pre-
The quantity for surveying the pixel component of sub- MB determines second bit number.
9. the method according to the description of claim 7 is characterized in that the prediction residual and the first bit that pass through the MB to be predicted
Several the second prediction residuals that the MB to be predicted is determined with the prediction residual of the sub- MB to be predicted and the second bit number, comprising:
The first weighted value of the MB to be predicted is determined according to the prediction residual of the MB to be predicted and the first bit number；
The second weighted value of the MB to be predicted is determined according to the prediction residual of the sub- MB to be predicted and the second bit number；
First weighted value and second weighted value are judged；
If first weighted value is greater than second weighted value, continue to divide the sub- MB to be predicted；
If first weighted value is less than second weighted value, choose described in the prediction residual conduct of the sub- MB to be predicted
The second prediction residual of MB to be predicted.
10. the method according to claim 1, wherein according to first residual error it is subjective and with it is described second residual
Final prediction technique that is poor subjective and determining the MB to be predicted, comprising:
Choose first residual error it is subjective and with the second residual error subjectivity and in minimum value, institute is determined by the minimum value
State the final prediction technique of MB to be predicted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811260616.7A CN109561303B (en) | 2018-10-26 | 2018-10-26 | Prediction method based on video compression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811260616.7A CN109561303B (en) | 2018-10-26 | 2018-10-26 | Prediction method based on video compression |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109561303A true CN109561303A (en) | 2019-04-02 |
| CN109561303B CN109561303B (en) | 2020-12-15 |
Family
ID=65865385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811260616.7A Active CN109561303B (en) | 2018-10-26 | 2018-10-26 | Prediction method based on video compression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109561303B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102281434A (en) * | 2010-06-10 | 2011-12-14 | 中国移动通信集团公司 | Video compression method and equipment |
| CN102595135A (en) * | 2012-02-24 | 2012-07-18 | 中国科学技术大学 | Method and device for scalable video coding |
| CN102917197A (en) * | 2011-08-04 | 2013-02-06 | 想象技术有限公司 | External vectors in a motion estimation system |
| CN103517069A (en) * | 2013-09-25 | 2014-01-15 | 北京航空航天大学 | HEVC intra-frame prediction quick mode selection method based on texture analysis |
| US20140205015A1 (en) * | 2011-08-25 | 2014-07-24 | Telefonaktiebolaget L M Ericsson (Publ) | Depth Map Encoding and Decoding |
| CN106034235A (en) * | 2015-03-11 | 2016-10-19 | 杭州海康威视数字技术股份有限公司 | Method for calculating coding distortion degree and coding mode control and system thereof |
| CN107509076A (en) * | 2017-08-25 | 2017-12-22 | 中国软件与技术服务股份有限公司 | A kind of Encoding Optimization towards ultra high-definition video |
-
2018
- 2018-10-26 CN CN201811260616.7A patent/CN109561303B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102281434A (en) * | 2010-06-10 | 2011-12-14 | 中国移动通信集团公司 | Video compression method and equipment |
| CN102917197A (en) * | 2011-08-04 | 2013-02-06 | 想象技术有限公司 | External vectors in a motion estimation system |
| US20140205015A1 (en) * | 2011-08-25 | 2014-07-24 | Telefonaktiebolaget L M Ericsson (Publ) | Depth Map Encoding and Decoding |
| CN102595135A (en) * | 2012-02-24 | 2012-07-18 | 中国科学技术大学 | Method and device for scalable video coding |
| CN103517069A (en) * | 2013-09-25 | 2014-01-15 | 北京航空航天大学 | HEVC intra-frame prediction quick mode selection method based on texture analysis |
| CN106034235A (en) * | 2015-03-11 | 2016-10-19 | 杭州海康威视数字技术股份有限公司 | Method for calculating coding distortion degree and coding mode control and system thereof |
| CN107509076A (en) * | 2017-08-25 | 2017-12-22 | 中国软件与技术服务股份有限公司 | A kind of Encoding Optimization towards ultra high-definition video |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109561303B (en) | 2020-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104935938B (en) | Inter-frame prediction method in a kind of hybrid video coding standard | |
| RU2544799C2 (en) | Moving image encoding device, moving image decoding device, moving image encoding method and moving image decoding method | |
| US20140126635A1 (en) | Optimal intra prediction in block-based video coding | |
| CN105430415B (en) | Fast encoding method in a kind of 3D HEVC deep video frames | |
| CN103248895B (en) | A kind of quick mode method of estimation for HEVC intraframe coding | |
| TWI690199B (en) | Video coding intra prediction method and device | |
| KR20110134319A (en) | Methods for encoding/decoding high definition image and apparatuses for performing the same | |
| CN105959698A (en) | Method and apparatus for performing interpolation based on transform and inverse transform | |
| CN108259898A (en) | Fast encoding method in frame based on Quality scalable video coding QSHVC | |
| CN113507609B (en) | Interframe image parallel coding method based on time-space domain prediction | |
| CN109688411B (en) | Video coding rate distortion cost estimation method and device | |
| CN109561303A (en) | A kind of prediction technique based on video compress | |
| CN109547788B (en) | Image compression method, equipment and image transmission system | |
| CN109474799A (en) | Image storage method and its system based on video monitoring | |
| CN105809717A (en) | Depth estimation method, system and electronic equipment | |
| CN109379599A (en) | Complex texture prediction technique based on Video coding | |
| CN110049339A (en) | Prediction direction choosing method, device and storage medium in image coding | |
| CN109587481B (en) | Video encoding method and apparatus | |
| CN113055678A (en) | Measurement domain compressed sensing coding algorithm based on adjacent pixel correlation | |
| CN109510995A (en) | A kind of prediction technique based on video compress | |
| Hu et al. | Low-complexity progressive image transmission scheme based on quadtree segmentation | |
| CN109660793B (en) | Prediction method for bandwidth compression | |
| CN109561301A (en) | A kind of prediction technique in video compress | |
| CN109327701A (en) | Complex texture adaptive forecasting method in bandwidth reduction | |
| JP5360614B2 (en) | AC component prediction method and image processing apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20201127 Address after: No. 02, 1701, No. 17, Zhujiang West Road, Tianhe District, Guangzhou City, Guangdong Province Applicant after: GUANGDONG HONGSHI DIGITAL MEDIA Co.,Ltd. Address before: 710065 No. 86 Leading Times Square (Block B), No. 2, Building No. 1, Unit 22, Room 12202, No. 51, High-tech Road, Xi'an High-tech Zone, Shaanxi Province Applicant before: Xi'an Cresun Innovation Technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |