CN104243985B - Sample adaptive equalization in HEVC - Google Patents
Sample adaptive equalization in HEVC Download PDFInfo
- Publication number
- CN104243985B CN104243985B CN201410510122.5A CN201410510122A CN104243985B CN 104243985 B CN104243985 B CN 104243985B CN 201410510122 A CN201410510122 A CN 201410510122A CN 104243985 B CN104243985 B CN 104243985B
- Authority
- CN
- China
- Prior art keywords
- pixel
- types
- type
- compensation
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
Sample adaptive equalization in HEVC is improved.First, extra boundary compensation is carried out for the boundary pixel of each coding unit in maximum coding unit.Secondly, new pattern is added in edge compensation (EO) type 2 and 3：Current pixel differs 1 with the pixel value of left side adjacent pixel, and more than 3 are differed with right side adjacent pixel, and current pixel differs 2 with the pixel value of left side adjacent pixel, and more than 6 are differed with right side adjacent pixel.Finally, the situation that the difference of current pixel and adjacent pixel in type 2 and 3 is less than or equal to 1 is eliminated.
Description
Joint study
The application obtains following fund assistance by North China University of Tech and the joint study of information institute of Beijing Jiaotong University：
State natural sciences fund (No.60903066, No.60972085), Beijing's Natural Science Fund In The Light (No.4102049), education
Portion's beginning teacher's fund (No.20090009120006), Beijing institution of higher education talent directly under the jurisdiction of a municipal government teach in-depth to plan by force
(PHR201008187)。
Technical field
The present invention relates to image processing field, more specifically, is related to the sample adaptive equalization of the optimization in HEVC.
Background technology
In April, 2010, two big international video encoding standard tissue VCEG and MPEG set up video compress joint group JCT-
VC (Joint collaborative Team on Video Coding), together develops efficient video coding HEVC (High
Efficiency video coding) standard, it is also referred to as H.265.HEVC standard main target is and previous generation standards
H.264/AVC the raising of significantly code efficiency is realized, in particular for high-resolution video sequence.Its target is identical
Code check is reduced to the 50% of H.264 standard under video quality (PSNR).
With regard to the current stage, HEVC still continues to use the hybrid encoding frame for H.264 beginning to use.Interframe and infra-frame prediction are compiled
Code：Eliminate the correlation of time-domain and spatial domain.Transition coding：Transition coding is carried out to residual error to eliminate spatial coherence.Entropy
Coding：Eliminate statistical redundancy.HEVC will put forth effort to study new coding tools or technology, carry in hybrid encoding frame
High video compression efficiency.
At present, the new features of many codings proposed in the discussion of JCT-VC tissues, it is possible to HEVC marks can be added
In standard, the specific document of each discussion can be fromhttp：//wftp3.itu.intObtain.
HEVC (High Efficiency Video Coding) standard is formally used as international standard in the January, 2013
Issue.It is referred to as H.265 in ITU-T, and the Part II of MPEG-H is referred to as in ISO/IEC.Meanwhile in order to make HEVC
Standard can support more application scenarios, carry out follow-up work, including to high-resolution and color format
Support, scalable coding, 3-D/ is three-dimensional/multi-view coded.
But during HEVC is still continued to develop.Sample adaptive equalization is a fine emerging new technology in HEVC, should
Technology adds new blood to HEVC, but the technology also there is some shortcomings and needs to mend due to being emerging
The place filled, it is described in detail below.
In the present invention, it is proposed that two kinds of sample adaptive equalization optimization algorithms, from its experimental result we
It can be seen that both algorithms suffer from good effect.
In the description of the invention, following document is refer to, the present invention is incorporated into herein and is used as disclosure of the invention
A part for content.
[1] .G.J.Sullivan, J.R.Ohm, W.J.Han and T.Wiegand.Overview of the High
Efficiency Video Coding (HEVC) Standard, IEEE Transactions on Circuits and
Systems for Video Technology, Vol.22, Issue 12,2012, pp.1649-1668.
[2] .J.Zhu, D.Zhou, G.He, S.Goto.A combined SAO and de-blocking filter
Architecture for HEVC video decoder, in Proceedings of the Conference on ' 20th
IEEE International Conference on Image Processing (ICIP) ', Melbourne, VIC, 15-
18Sept.2013, pp.1967-1971.
[3] .P.N.Subramanya, R.Adireddy, D.Anand.SAO in CTU decoding loop for
HEVC video decoder, in Proceedings of the Conference on ' International
Conference on Signal Processing and Communication (ICSC) ', Noida, 12-14Dec.2013,
pp.507-511.
[4] .G.B.Praveen, G.B.Praveen.Analysis and approximation of SAO
Estimation for CTU-level HEVC encoder, in Proceedings of the Conference on '
Visual Communications and Image Processing (VCIP) ', Kuching, 17-20Nov.2013, pp.1-
5.
[5] .W.Kim, J.Sole, Marta.Offset Scaling in SAO for High Bit-depth Video
Coding, in Proceedings of the Conference on ' 13th Joint Collaborative Team on
Video Coding (JCT-VC) Meeting ', Incheon, KR, 18-26Apr.2013, JCTVC-M0335.
[6] .G.Laroche, T.Poirier, C.Gisquet, E.P.Onno.On Inter-Layer SAO
For Base mode, in Proceedings of the Conference on ' 12th Joint Collaborative
Team on Video Coding (JCT-VC) Meeting ', Geneva, CH, 14-23Jan.2013, JCTVC-L107.
The content of the invention
In order to solve the above-mentioned technical problem, this patent is improved the sample adaptive equalization in HEVC.First, for
The boundary pixel of each coding unit in maximum coding unit carries out extra boundary compensation.Secondly, at edge compensation (EO)
New pattern is added in type 2 and 3：Current pixel differs 1 with the pixel value of left side adjacent pixel, is differed with right side adjacent pixel
More than 3, current pixel differs 2 with the pixel value of left side adjacent pixel, and more than 6 are differed with right side adjacent pixel.Finally, eliminate
The difference of current pixel and adjacent pixel is less than or equal to 1 situation in type 2 and 3.
According to one side, a kind of method for being used to determine the sample adaptive equalization parameter of decoder end in HEVC,
The method is implemented in the encoder, the described method includes：
Maximum coding unit (CU) is rebuild from encoded image；
It is based respectively on following sample adaptive equalization type to carry out sample adaptive equalization to the maximum CU of reconstruction：Pixel
Band compensation (Band offset, BO) type, edge compensation (Edge offset, EO) Class1, EO types 2, EO types 3 and EO
Type 4, so as to be compensated for the BO types as a result, and being that EO Class1, EO types 2, EO types 3 and EO types 4 obtain 4
A compensation result；
4 compensation knots of compensation result and EO Class1, EO types 2, EO types 3 and EO types 4 based on the BO types
The maximum CU of each and the reconstruction in fruit relative to original image rate distortion costs value, from BO types, EO types 0,
Optimal sample adaptive equalization type is selected in EO Class1, EO types 2, EO types 3 and EO types 4, wherein, the EO types 0
Represent without compensation and thus corresponding to the maximum CU of the reconstruction；
Definite and selected corresponding first offset of optimal sample adaptive equalization type, wherein, the EO classes
Corresponding first offset of type 0 is 0；
The second boundary offset is determined based on selected optimal sample adaptive equalization type,
Selected optimal sample adaptive equalization type, first offset and the second boundary offset are compiled
Transmitted in code to the coding information associated with the maximum CU；
It is characterized in that：
The pixel value comprising current pixel is smaller than the pixel value of left side adjacent pixel by 1 in type of pixel 2, picture more adjacent than right side
The situation of the pixel value big more than 3 of element, and the pixel value of current pixel are smaller than the pixel value of left side adjacent pixel by 2, than right side
The situation of the pixel value big more than 6 of adjacent pixel,
The pixel value comprising current pixel is bigger than the pixel value of right side adjacent pixel by 1 in type of pixel 3, picture more adjacent than left side
The situation of the pixel value small more than 3 of element, and the pixel value of current pixel are bigger than the pixel value of right side adjacent pixel by 2, than left side
The situation of the pixel value small more than 6 of adjacent pixel,
In type of pixel 2 and type of pixel 3, if the pixel value of the pixel value of current pixel and two adjacent pixels it
Between difference be less than or equal to 1, then current pixel do not compensate,
The definite the second boundary offset includes：Based on the optimal sample adaptive equalization type, calculating is directed to institute
Multiple rate distortion costs values of multiple boundary compensation values of the boundary pixel of each converter unit in maximum CU are stated, and are based on institute
State multiple rate distortion costs values and select a boundary compensation value.
In further, the grammer member in the maximum CU is set based on optimal sample adaptive equalization type
The value of plain sao_type_idx and sao_eo_class, wherein, the value 0,1,2 of sao_type_idx is represented without sample respectively
Adaptive equalization, use pixel band compensation (BO) and edge compensation (EO), value 0-3 expression EO Class1, the EO of sao_eo_class
Type 2, EO types 3 and EO types 4.
In further, the EO Class1-4 is respectively using horizontal line, vertical line, 135 degree of angle diagonal and 45
Spend angle diagonal.
In further, in the BO types, both full-pixel scope is uniformly divided into 32 pixel bands, is selected
Wherein 4 pixel bands, and determine the positive compensation for 4 selected pixel bands or negative compensation, wherein, use rate distortion costs
To determine the positive compensation for 4 selected pixel bands or negative compensation, the optimal sample adaptive equalization type is described
BO types, first offset include determining the positive compensation for 4 selected pixel bands or negative compensation.
In further, in the EO Class1-4, first offset includes being based on and the EO types
Offset determined by the corresponding penalty functions of 1-4.
In further, in the EO Class1-4, first offset includes being based on and the EO types
Offset determined by the corresponding penalty functions of 1-4.
According to another aspect, a kind of device for being used to determine the sample adaptive equalization parameter of decoder end in HEVC,
Described device is implemented in the encoder, and described device includes：
For rebuilding the unit of maximum coding unit (CU) from encoded image；
Adaptively mended for being based respectively on following 4 kinds of sample adaptive equalization types to carry out sample to the maximum CU of reconstruction
Repay：Pixel band compensation (Band offset, BO) type, edge compensation (Edge offset, EO) Class1, EO types 2, EO classes
Type 3 and EO types 4, so as to be compensated for the BO types as a result, and being EO Class1, EO types 2, EO types 3 and EO types
4 obtain the unit of 4 compensation results；
4 benefits for the compensation result based on the BO types and EO Class1, EO types 2, EO types 3 and EO types 4
The maximum CU of each and the reconstruction in result is repaid relative to the rate distortion costs value of original image, from BO types, EO classes
Optimal sample adaptive equalization type is selected in type 0, EO Class1, EO types 2, EO types 3 and EO types 4, wherein, the EO
Type 0 represents the unit without compensation and the maximum CU thus corresponding to the reconstruction；
For determining the unit with selected corresponding first offset of optimal sample adaptive equalization type, its
In, corresponding first offset of the EO types 0 is 0；
For determining the unit of the second boundary offset based on selected optimal sample adaptive equalization type,
For selected optimal sample adaptive equalization type, first offset and the second boundary to be compensated
The unit that value coding is transmitted into the coding information associated with the maximum CU；
It is characterized in that：
The pixel value comprising current pixel is smaller than the pixel value of left side adjacent pixel by 1 in type of pixel 2, picture more adjacent than right side
The situation of the pixel value big more than 3 of element, and the pixel value of current pixel are smaller than the pixel value of left side adjacent pixel by 2, than right side
The situation of the pixel value big more than 6 of adjacent pixel,
The pixel value comprising current pixel is bigger than the pixel value of right side adjacent pixel by 1 in type of pixel 3, picture more adjacent than left side
The situation of the pixel value small more than 3 of element, and the pixel value of current pixel are bigger than the pixel value of right side adjacent pixel by 2, than left side
The situation of the pixel value small more than 6 of adjacent pixel,
In type of pixel 2 and type of pixel 3, if the pixel value of the pixel value of current pixel and two adjacent pixels it
Between difference be less than or equal to 1, then current pixel do not compensate,
It is described to be used to determine that the unit of the second boundary offset includes：For based on the optimal sample adaptive equalization class
Type, calculates multiple rate distortion generations of multiple boundary compensation values of the boundary pixel for each converter unit being directed in the maximum CU
It is worth, and the unit of a boundary compensation value is selected based on the multiple rate distortion costs value.
According on the other hand, the present invention proposes a kind of computer program product, described it includes computer-readable medium
Computer-readable medium includes program code, and said program code when being executed by a processor, performs method as described above.
According on the other hand, the present invention proposes a kind of codec based on HEVC, it is used to perform as described above
Method for video coding.
Brief description of the drawings
Fig. 1 shows the high level flow chart of sample adaptive equalization according to an embodiment of the invention.
Fig. 2 shows the schematic diagram of pixel band compensation according to an embodiment of the invention.
Fig. 3 shows pixel edge compensation type judgement according to an embodiment of the invention.
Fig. 4 shows the type schematic diagram of similar edge compensation according to an embodiment of the invention.
Fig. 5 shows the edge cells (dotted portion inside maximum coding unit according to an embodiment of the invention
For converter unit boundary pixel).
Fig. 6 shows the schematic diagram of second and third class sample adaptive equalization according to an embodiment of the invention.
Fig. 7 shows the signal of the type of pixel according to an embodiment of the invention for not carrying out sample adaptive equalization
Figure.
Fig. 8 shows the request according to an embodiment of the invention that need not carry out sample adaptive equalization.
Fig. 9 (a) and (b) show the supplement voxel model of type 2 according to an embodiment of the invention.
Fig. 9 (c) and (d) show the supplement voxel model of type 3 according to an embodiment of the invention.
Fig. 9 (e) and (f) show what needs according to an embodiment of the invention were rejected from type 2 and 3
Voxel model.
Figure 10 shows method schematic according to an embodiment of the invention.
Figure 11 (a) shows method flow diagram according to an embodiment of the invention.
Figure 11 (b) shows device flow chart according to an embodiment of the invention.
Figure 12 shows the basic principle figure of HEVC encoders.
Embodiment
Various schemes are described referring now to attached drawing.In the following description, in order to explain, elaborate multiple specific thin
Section is in order to provide the thorough understanding to one or more schemes.It may be evident, however, that also can in the case of these no details
Enough realize these schemes.
As used in this specification, term " component ", " module ", " system " etc. are intended to refer to related to computer
Entity, such as, but not limited to, hardware, firmware, the combination of hardware and software, software, or executory software.For example,
Component can be but not limited to：The process run on a processor, processor, object, executable (executable), perform
Thread, program, and/or computer.For example, running application program on the computing device and the computing device can be
Component.One or more assemblies can be located in executive process and/or execution thread, and component can be located at a calculating
On machine and/or it is distributed on two or more platform computers.In addition, these components can be from various with what is be stored thereon
The various computer-readable mediums of data structure perform.Component can be communicated by means of locally and/or remotely process, such as
According to the signal with one or more packets, for example, coming from by means of in signal and local system, distributed system
Another component interaction and/or one with being interacted on the network of such as internet etc by means of signal with other systems
The data of component.
Figure 12 shows the general configuration figure for the video encoder that efficient video coding (HEVC) is realized.The coding of HEVC
Device framework is roughly the same with H.264 used encoder architecture, is carried out primarily directed to the algorithm used in modules
Further research, improvement, in particular for high-resolution video sequence, its improved target is in same video quality
(PSNR) code check is reduced to the 50% of H.264 standard under.
Since the encoder architecture of HEVC is roughly the same with H.264 used encoder architecture, this hair is not obscured
It is bright, the overall architecture in Fig. 9 is not described in the application.
In HEVC, the video of input is first separated into some fritters and is called code tree unit (coding tree
Units, CTU).It will be understood by those skilled in the art that concepts of the CTU equivalent to the macro block (macroblock) of standard before.
Coding unit (CU) is square (pixel) unit for possessing a prediction mode (in frame, interframe or skip).Base
It is as shown in Figure 1 in the predicting unit partitioning scheme of CTU and CU.
1.1 sample adaptive equalization brief introductions
Sample adaptive equalization is that a fine emerging new technology, the technology add fresh blood to HEVC in HEVC
Liquid.
New loop filter, including de-blocking filter (Deblocking Filter, DF) and sample are employed among HEVC
This adaptive equalization (Sample Adaptive Offset, SAO).Sample adaptive equalization is located at after de-blocking filter, mesh
Be the average distortion measure for reducing region, sample adaptive equalization can use selected sorting technique by maximum coding unit first
Sample carry out statistic of classification, so as to obtain obtaining an offset for various species, offset is then added to corresponding species
Pixel value in.
Adaptive sample compensation is a module after being located at de-blocking filter during coding, which completes
Be one modification decoding sample value a process, modification process is value [1] certain to sample compensation in any case
[2].Sample adaptive equalization filtering is acted on certain area, can all pass through a sao_ in each maximum coding unit
The syntax elements of type_idx choose the filter patterns of the unit, wherein " sao " then represents sample adaptive equalization
(samples adaptive offset).When sao_type_idx for 0 is that to represent that the maximum coding unit does not apply to sample adaptive
Should compensate, when the value is 1 and 2 it is corresponding represent using pixel band compensate (BO, band offset) and edge compensation (EO,
edge offset)[3][4].Sample adaptive equalization flow chart is as shown in Figure 1.
In pixel band compensation model, the offset of selection depends directly on the amplitude of pixel.In this mode, full figure
What the scope of element was unified is divided into 32 parts, that is, divide into 32 pixel bands, as shown in Fig. 2, pixel band model meeting
Wherein continuous 4 pixel bands are chosen, and the sample value in this four pixel bands is carried out just (to) to compensate or bear (to) compensation.
Why to choose continuous 4 pixel bands and compensate is to be usually found in because of band artifacts in smooth region, and
Pixel value in one maximum coding unit is intended to concentrate on less pixel band.4 offsets and edge are chosen at the same time
The operation of compensation model matches, because edge compensation is also that have chosen 4 offsets.
In edge compensation, the value of syntax elements sao_eo_class represents to use water in edge compensation from 0 to 3
Flat, vertical, 135 degree of angle diagonal and 45 degree of angle diagonal, as shown in Figure 3.All pixel values all can in maximum coding unit
One be classified as in 5 types as shown in table 1 and Fig. 4, wherein c represent current pixel value, and a and b represent adjacent respectively
Two pixel values.This foundation sorted out is to have decoded each pixel sample values come, so and being not required extra letter
Number go to transmit this categorization results [5].The edge index classification in pixel samples is relied on, decoding end is transferred to by one
Look-up table, we can add corresponding offset to corresponding sample value.For species 1 and 2, offset is always just (to)
, and the offset of species 3 and 4 always bears (to), therefore be a smooth effect caused by edge filter.
In Fig. 4, what species 1,2 carried out is positive compensation, and what species 3,4 carried out is negative sense compensation.
The compensation classification judgement of 1 pixel boundary of table
| Species | Condition |
| 1 | C ＜ a＆＆c ＜ b |
| 2 | (c ＜ a＆＆c==b) | | (c==a＆＆c ＜ b) |
| 3 | (c ＞ a＆＆c==b) | | (c==a＆＆c ＞ b) |
| 4 | C ＞ a＆＆c ＞ b |
| 0 | Non- case above |
In conclusion the pixel band compensation of sample adaptive equalization and edge compensation all can be respectively that each maximum coding is single
Member transmits the amplitude of 4 offsets, and is compensated for pixel band, and the symbol (positive or negative) of compensation is [6] for needing to be transmitted.
The selection of offset and corresponding syntax elements value, such as sao_type_idx and sao_eo_class, are all to pass through optimization rate
The effect of distortion is chosen.Meanwhile the sample adaptive equalization parameter of current block can also be by inheriting the left side and top
The sample adaptive equalization parameter of maximum coding unit obtains.On the whole, sample adaptive equalization, which is one, allows into one
Step improves the non-linear filtering operation of reconstruction signal, and he can not only strengthen the signal representation of smooth region but also can improve side
Edge point.
2. the sample adaptive equalization optimization based on converter unit border
The size of converter unit (TransformUnit, TU) in HEVC is from 4x4 to 32x32, and is become in h .264
The size for changing unit is not more than 8x8.Because it can be drawn using discrete cosine transform (Discrete Cosine Transform, DCT)
The quantizing distortion of conversion coefficient is played, so the converter unit of bigger can cause more obvious annular pseudo- at the edge of converter unit
Shadow and more serious distortion.But sample adaptive equalization is accomplished that the reduction of the average distortion measure in region, block elimination filtering
Device does not have the effect for reducing the distortion factor, therefore the sample distortion for TU edges is not handled particularly.
Optimize algorithm
Found by our research, even the converter unit after have passed through de-blocking filter, converter unit border
On distortion it is also more much larger than inside converter unit.In order to targetedly reduce the borderline distortion of converter unit, we
A kind of new pattern, i.e. converter unit edge pattern are added for sample adaptive equalization.The pattern is intended to give converter unit side
The extra compensation in one, boundary, so as to bring lower converter unit boundary sample distortion and lower bit rate.Maximum coding is single
First inner transformation unit is as shown in figure 5, dotted portion therein is the borderline pixel samples of converter unit.
By statistics we have found that the distortion of converter unit boundary is bigger by 20% or so than inside converter unit, in order into
One step obtains the difference of compensation required for inside converter unit border and converter unit, we are by converter unit boundary sample and change
Change unit internal specimen and pass through sample adaptive equalization.The results show that for edge compensation, converter unit border is averaged
The distortion factor wants big by 0.45, is compensated for pixel band, and the average distortion measure on converter unit border wants big by 0.4.
Thus, we have proposed following novel compensation way：
A) when being counted to sample, by the pixel samples inside converter unit and converter unit borderline sample point
Do not counted.
B) optimal compensation model and offset are calculated according to original sample adaptive equalization, calculates extracting rate distortion generation
Valency.As shown in the types index 0 of table 2, i.e., extra compensation is not carried out to converter unit border.
C) extra compensation is carried out to the border of converter unit, types index 1,2,3 to converter unit side in traversal list 2
Boundary carries out the strategy of additional compensation, i.e., compensates converter unit again after converter unit internal compensation value being added the value in table 2
It is internal.Calculate rate distortion costs caused by these three types.
D) four more above-mentioned rate distortion costs values, select the compensation type of the type current block the most of Least-cost,
And index value is transmitted.
Additional compensation of the table 2 for converter unit
Test result indicates that this algorithm compares former algorithm reduces -0.13 by BD-rate, sample adaptive compensation algorithm
Performance realize 26% lifting, it can be seen that this algorithm has certain effect for the optimization of sample adaptive equalization.
3. the compensation model optimization of sample adaptive equalization
In the course of the study it was found that requirement of the sample adaptive equalization for sample statistics is very harsh, deposit
Corresponding compensation is arrived what substantial amounts of pixel that is regular, needing to compensate did not had.As shown in fig. 6, in second and third class
In sample adaptive equalization, it may be seen that only when current pixel have to it is complete with the pixel value of an adjacent pixel
It can just judge whether current pixel belongs to second and third class sample adaptive equalization when identical.Caused by such compensation policy
Problem is to exist substantial amounts of as shown in fig. 7, i.e. pixel gradient cannot but be judged as greatly carrying out the picture of sample adaptive equalization very much
Element, does not compensate but, and pixel c only differs very little with pixel a or b in Fig. 7, but gradient is but very big, and such case is complete
Meeting sample adaptive equalization causes the more smooth design original intention in marginal portion, and pixel and Fig. 6 types in Fig. 7 entirely
Pixel difference only in 1,2 pixel value, therefore we are based on this idea, it is proposed that this algorithm.
At the same time, it has been found that also there is a kind of unreasonable situation in sample adaptive equalization.Such case is：
Current pixel is judged as second or the third type, even if but current pixel value and adjacent pixel only differ one or
Seldom pixel value, i.e. pixel gradient very little, as shown in figure 8, the position can also be compensated according to second or the third type.This
Clearly unreasonable, because current pixel is very smooth, the space not further smoothed, it is adaptive to have run counter to sample
The original intention compensated to edge that should be compensated.Therefore, one of content that we also study this point as us.
Therefore, it is proposed that tetra- kinds of patterns of a, b, c, d as shown in Figure 9 are added in sample adaptive equalization can will very
The pixel samples required supplementation with without compensating are brought into sample adaptive equalization more, and remove e, f both of which more
Added with beneficial to obtaining more accurately offset.
Specifically, the situation as shown in Fig. 9 (a) is that current pixel c differs 1 with the pixel value of adjacent pixel a, and with
The pixel value difference more than 3 of the adjacent pixel b of another side；Situation shown in Fig. 9 (b) is current pixel c and adjacent pixel a
Pixel value difference 2, and differ more than 6 with the pixel value of the pixel b of another side；Situation as shown in Fig. 9 (c) is current pixel
C differs 1 with the pixel value of adjacent pixel b, and differs more than 3 with the pixel value of the adjacent pixel a of another side；Shown in Fig. 9 (d)
Situation be that current pixel c differs 2 with the pixel value of adjacent pixel b, and differ more than 6 with the pixel value of the pixel a of another side.
We bring the situation of Fig. 9 (a), (b) into type 2, and Fig. 9 (c), the situation of (d) are brought into type 3, afterwards
The situation of Fig. 9 (e), (f) are eliminated the 2nd of sample adaptive equalization the, 3 classes and compensated by us.
Then counted according to sample adaptive equalization standard step and rate-distortion optimization, select current maximum coding
Sample adaptive equalization is carried out to current maximum coding unit after the optimal compensation model of unit.
Figure 11 (a) shows a kind of sample for being used to determine decoder end in HEVC according to an embodiment of the invention
The flow chart of the method for this adaptive equalization parameter, this method are implemented in the encoder.
According to the schematic diagram shown in Figure 10：
In step 1101, in the encoder, maximum coding unit (CU) is rebuild from encoded image.
In step 1102, it is based respectively on following 4 kinds of samples adaptive equalization type to carry out sample to the maximum CU of reconstruction
This adaptive equalization：Pixel band compensation (Band offset, BO) type, edge compensation (Edge offset, EO) Class1, EO
Type 2, EO types 3 and EO types 4, so as to be compensated for the BO types as a result, and being EO Class1, EO types 2, EO classes
Type 3 and EO types 4 obtain 4 compensation results.
In step 1103, compensation result and EO Class1, EO types 2, EO types 3 and EO classes based on the BO types
The maximum CU of each and the reconstruction in 4 compensation results of type 4 relative to original image rate distortion costs value, from BO
Optimal sample adaptive equalization type is selected in type, EO types 0, EO Class1, EO types 2, EO types 3 and EO types 4, its
In, the EO types 0 are represented without compensation and thus corresponding to the maximum CU of the reconstruction.
In step 1104, definite and selected corresponding first offset of optimal sample adaptive equalization type,
Wherein, corresponding first offset of the EO types 0 is 0.
In step 1105, the second boundary offset is determined based on selected optimal sample adaptive equalization type.
In a step 1106, by selected optimal sample adaptive equalization type, first offset and described
Two boundary compensation values are encoded into the coding information associated with the maximum CU and transmitted.
In one embodiment of the invention, the pixel value of current pixel is included in pixel 2 than left side adjacent pixel
Pixel value is small by 1, bigger than the pixel value of right side adjacent pixel more than 3 situation, and the pixel value of current pixel is more adjacent than left side
The pixel value of pixel is small by 2, bigger than the pixel value of right side adjacent pixel more than 6 situation.
In one embodiment of the invention, picture of the pixel value comprising current pixel than right side adjacent pixel in pixel 3
Element is worth big by 1, smaller than the pixel value of left side adjacent pixel more than 3 situation, and the pixel value picture more adjacent than right side of current pixel
The pixel value of element is big by 2, smaller than the pixel value of left side adjacent pixel more than 6 situation.
In one embodiment of the invention, in type of pixel 2 and type of pixel 3, if the pixel value of current pixel
Difference is less than or equal to 1 between the pixel value of two adjacent pixels, then current pixel does not compensate.
In step 1105, determine that the second boundary offset further comprises：Based on the optimal sample adaptive equalization
Type, calculates multiple rate distortions of multiple boundary compensation values of the boundary pixel for each converter unit being directed in the maximum CU
Cost value, and a boundary compensation value is selected based on the multiple rate distortion costs value.
In a preferred embodiment of the invention, the maximum CU is set based on optimal sample adaptive equalization type
In syntactic element sao_type_idx and sao_eo_class value, wherein, the value 0,1,2 of sao_type_idx represents respectively
Represented without sample adaptive equalization, using pixel band compensation (BO) and edge compensation (EO), the value 0-3 of sao_eo_class
EO Class1, EO types 2, EO types 3 and EO types 4.
In a preferred embodiment of the invention, the EO Class1-4 is respectively using horizontal line, vertical line, 135 degree of angles
Diagonal and 45 degree of angle diagonal.
In a preferred embodiment of the invention, in the BO types, both full-pixel scope is uniformly divided into 32
Pixel band, selects wherein 4 pixel bands, and determines the positive compensation for 4 selected pixel bands or negative compensation, wherein, make
Determine adaptively to mend for the positive compensation of 4 selected pixel bands or negative compensation, the optimal sample with rate distortion costs
It is the BO types to repay type, and first offset includes determining the positive compensation or negative for 4 selected pixel bands
Compensation.
In a preferred embodiment of the invention, in the EO Class1-4, first offset include be based on
Offset determined by -4 corresponding penalty function of EO Class1.
In a preferred embodiment of the invention, in the EO Class1-4, first offset include be based on
Offset determined by -4 corresponding penalty function of EO Class1.
Figure 11 (b) shows a kind of sample for being used to determine decoder end in HEVC according to an embodiment of the invention
The block diagram of the device of this adaptive equalization parameter, the device are implemented in the encoder.Unit in Figure 11 (b) is right respectively
Should be in the correlation method step in Figure 11 (a), therefore details are not described herein.
The method disclosed in the present can be realized with software, hardware, firmware etc..
When implemented in hardware, video encoder can use general processor, digital signal processor (DSP), special collection
Into circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device
Part, discrete hardware components or its any combination for being designed as performing function described herein, to realize or perform.General processor
Can be microprocessor, but alternatively, which can also be any conventional processor, controller, microcontroller
Or state machine.Processor can also be embodied as the combination of computing device, for example, the combining of DSP and microprocessor, multiple micro- places
Manage the combination of device, the combination of one or more microprocessors and DSP core or any other such structure.In addition, at least one
A processor can perform one or more modules of above-mentioned one or more steps and/or operation including operable.
When with ASIC, FPGA when hardware circuit to realize video encoder when, its can include be configured as performing it is various
The various circuit blocks of function.Those skilled in the art can be according to the various constraintss applied over the whole system come with various
Mode designs and realizes these circuits, to realize various functions disclosed in this invention.
Although foregoing open file discusses exemplary arrangement and/or embodiment, it should be noted that being wanted without departing substantially from by right
In the case of seeking the scheme of description and/or the scope of embodiment that book defines, many can be made herein and changed and modifications.And
And although the key element for the scheme and/or embodiment for describing or requiring in the singular, it is also contemplated that the feelings of plural number
Condition, is limited to odd number unless expressly stated.In addition, all or part of any scheme and/or embodiment can with it is any its
All or part of combined use of its scheme and/or embodiment, unless indicating different.
Claims (8)
1. a kind of method for being used to determine the sample adaptive equalization parameter of decoder end in HEVC, the method is in encoder
Middle implementation, the described method includes：
Maximum coding unit (CU) is rebuild from encoded image；
It is based respectively on following sample adaptive equalization type to carry out sample adaptive equalization to the maximum CU of reconstruction：Pixel band is mended
(BO) type, edge compensation (EO) Class1, EO types 2, EO types 3 and EO types 4 are repaid, so as to be mended for the BO types
Repay as a result, and obtaining 4 compensation results for EO Class1, EO types 2, EO types 3 and EO types 4；
Compensation result and EO Class1, EO types 2, EO types 3 based on the BO types and in 4 compensation results of EO types 4
Each and the reconstruction maximum CU relative to original image rate distortion costs value, from BO types, EO types 0, EO classes
Optimal sample adaptive equalization type is selected in type 1, EO types 2, EO types 3 and EO types 4, wherein, the EO types 0 represent
Without compensation and thus corresponding to the maximum CU of the reconstruction；
Definite and selected corresponding first offset of optimal sample adaptive equalization type, wherein, the EO types 0 are right
The first offset answered is 0；
The second boundary offset is determined based on selected optimal sample adaptive equalization type,
By selected optimal sample adaptive equalization type, first offset and the second boundary offset encode to
Transmitted in the coding information associated with the maximum CU；
It is characterized in that：
The pixel value comprising current pixel is smaller than the pixel value of left side adjacent pixel by 1 in type of pixel 2, than right side adjacent pixel
The situation of pixel value big more than 3, and the pixel value of current pixel are smaller than the pixel value of left side adjacent pixel by 2, more adjacent than right side
The situation of the pixel value big more than 6 of pixel,
The pixel value comprising current pixel is bigger than the pixel value of right side adjacent pixel by 1 in type of pixel 3, than left side adjacent pixel
The situation of pixel value small more than 3, and the pixel value of current pixel are bigger than the pixel value of right side adjacent pixel by 2, more adjacent than left side
The situation of the pixel value small more than 6 of pixel,
In type of pixel 2 and type of pixel 3, if poor between the pixel value of the pixel value of current pixel and two adjacent pixels
Less than or equal to 1, then current pixel does not compensate,
The definite the second boundary offset includes：Based on the optimal sample adaptive equalization type, calculate for described in most
Multiple rate distortion costs values of multiple boundary compensation values of the boundary pixel of each converter unit in big CU, and based on described more
A rate distortion costs value selects a boundary compensation value,
Wherein, set based on optimal sample adaptive equalization type syntactic element sao_type_idx in the maximum CU and
The value of sao_eo_class, wherein, the value 0,1,2 of sao_type_idx is represented without sample adaptive equalization, used respectively
Pixel band compensates (BO) and edge compensation (EO), and the value 0-3 of sao_eo_class represents EO Class1, EO types 2,3 and of EO types
EO types 4.
2. the method for claim 1, wherein the EO Class1-4 is respectively using horizontal line, vertical line, 135 degree of angles pair
Linea angulata and 45 degree of angle diagonal.
3. the method for claim 1, wherein in the BO types, both full-pixel scope is uniformly divided into 32 pictures
Plain band, selects wherein 4 pixel bands, and determines the positive compensation for 4 selected pixel bands or negative compensation, wherein, use
Rate distortion costs compensate to determine to be directed to the positive compensation of 4 selected pixel bands or bear, the optimal sample adaptive equalization
Type is the BO types, and first offset includes determining the positive compensation for being directed to 4 selected pixel bands or bear to mend
Repay.
4. the method for claim 1, wherein in the EO Class1-4, first offset includes being based on and institute
State offset determined by -4 corresponding penalty function of EO Class1.
5. a kind of device for being used to determine the sample adaptive equalization parameter of decoder end in HEVC, described device is in encoder
Middle implementation, described device include：
For rebuilding the unit of maximum coding unit (CU) from encoded image；
For being based respectively on following 4 kinds of samples adaptive equalization type to carry out sample adaptive equalization to the maximum CU of reconstruction：
Pixel band compensation (BO) type, edge compensation (EO) Class1, EO types 2, EO types 3 and EO types 4, so as to be the BO classes
Type is compensated as a result, and obtaining the unit of 4 compensation results for EO Class1, EO types 2, EO types 3 and EO types 4；
Tied for the compensation result based on the BO types and EO Class1,4 compensation of EO types 2, EO types 3 and EO types 4
The maximum CU of each and the reconstruction in fruit relative to original image rate distortion costs value, from BO types, EO types 0,
Optimal sample adaptive equalization type is selected in EO Class1, EO types 2, EO types 3 and EO types 4, wherein, the EO types 0
Represent the unit without compensation and the maximum CU thus corresponding to the reconstruction；
For determining the unit with selected corresponding first offset of optimal sample adaptive equalization type, wherein, institute
Corresponding first offset of EO types 0 is stated as 0；
For determining the unit of the second boundary offset based on selected optimal sample adaptive equalization type,
For selected optimal sample adaptive equalization type, first offset and the second boundary offset to be compiled
The unit transmitted in code to the coding information associated with the maximum CU；
It is characterized in that：
The pixel value comprising current pixel is smaller than the pixel value of left side adjacent pixel by 1 in type of pixel 2, than right side adjacent pixel
The situation of pixel value big more than 3, and the pixel value of current pixel are smaller than the pixel value of left side adjacent pixel by 2, more adjacent than right side
The situation of the pixel value big more than 6 of pixel,
The pixel value comprising current pixel is bigger than the pixel value of right side adjacent pixel by 1 in type of pixel 3, than left side adjacent pixel
The situation of pixel value small more than 3, and the pixel value of current pixel are bigger than the pixel value of right side adjacent pixel by 2, more adjacent than left side
The situation of the pixel value small more than 6 of pixel,
In type of pixel 2 and type of pixel 3, if poor between the pixel value of the pixel value of current pixel and two adjacent pixels
Less than or equal to 1, then current pixel does not compensate,
It is described to be used to determine that the unit of the second boundary offset includes：For based on the optimal sample adaptive equalization type,
Calculate multiple rate distortion costs of multiple boundary compensation values of the boundary pixel for each converter unit being directed in the maximum CU
It is worth, and the unit of a boundary compensation value is selected based on the multiple rate distortion costs value,
Wherein, set based on optimal sample adaptive equalization type syntactic element sao_type_idx in the maximum CU and
The value of sao_eo_class, wherein, the value 0,1,2 of sao_type_idx is represented without sample adaptive equalization, used respectively
Pixel band compensates (BO) and edge compensation (EO), and the value 0-3 of sao_eo_class represents EO Class1, EO types 2,3 and of EO types
EO types 4.
6. device as claimed in claim 5, wherein, in the BO types, both full-pixel scope is uniformly divided into 32 pictures
Plain band, selects wherein 4 pixel bands, and determines the positive compensation for 4 selected pixel bands or negative compensation, wherein, use
Rate distortion costs compensate to determine to be directed to the positive compensation of 4 selected pixel bands or bear, the optimal sample adaptive equalization
Type is the BO types, and first offset includes determining the positive compensation for being directed to 4 selected pixel bands or bear to mend
Repay.
7. a kind of computer-readable medium, the computer-readable medium includes program code, and said program code is when by handling
When device performs, the method as described in claim 1 is performed.
8. a kind of codec based on HEVC, it is used to perform such as claim 1-4 any one of them method for video coding.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410510122.5A CN104243985B (en) | 2014-09-29 | 2014-09-29 | Sample adaptive equalization in HEVC |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410510122.5A CN104243985B (en) | 2014-09-29 | 2014-09-29 | Sample adaptive equalization in HEVC |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104243985A CN104243985A (en) | 2014-12-24 |
| CN104243985B true CN104243985B (en) | 2018-05-08 |
Family
ID=52231207
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410510122.5A Active CN104243985B (en) | 2014-09-29 | 2014-09-29 | Sample adaptive equalization in HEVC |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104243985B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9877024B2 (en) * | 2015-03-06 | 2018-01-23 | Qualcomm Incorporated | Low complexity sample adaptive offset (SAO) coding |
| CN105306940B (en) * | 2015-10-27 | 2019-10-18 | 中星技术股份有限公司 | A kind of boundary compensation mode treatment method in coding and decoding video sample value migration |
| JP6634936B2 (en) * | 2016-04-05 | 2020-01-22 | 富士通株式会社 | Image encoding device, image encoding method, image encoding program, image decoding device, image decoding method, image decoding program, and image transmission system |
| CN110855973B (en) * | 2019-11-15 | 2021-10-26 | 北方工业大学 | Video intra-frame fast algorithm based on regional directional dispersion sum |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103209324A (en) * | 2012-01-17 | 2013-07-17 | 张民 | Method of applying edge offset |
| KR20130084052A (en) * | 2012-01-16 | 2013-07-24 | 주식회사 케이티 | Sample adaptive offset(sao) diagonal edge offset |
| KR20130084054A (en) * | 2012-01-16 | 2013-07-24 | 주식회사 케이티 | Sample adaptive offset (sao) edge offset |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9253482B2 (en) * | 2011-11-08 | 2016-02-02 | Texas Insturments Incorporated | Method and apparatus for sample adaptive offset without sign coding |
-
2014
- 2014-09-29 CN CN201410510122.5A patent/CN104243985B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130084052A (en) * | 2012-01-16 | 2013-07-24 | 주식회사 케이티 | Sample adaptive offset(sao) diagonal edge offset |
| KR20130084054A (en) * | 2012-01-16 | 2013-07-24 | 주식회사 케이티 | Sample adaptive offset (sao) edge offset |
| CN103209324A (en) * | 2012-01-17 | 2013-07-17 | 张民 | Method of applying edge offset |
Non-Patent Citations (2)
| Title |
|---|
| Low Complexity SAO in HEVC Base on Class Combination,Pre-decision and Merge Separation;Gaoxing Chen;《Proceedings of the 19th International Conference on Digital Signal Processing》;20140823;全文 * |
| Sample Adaptive Offset in the HEVC Standard;Chih-Ming Fu;《IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY》;20121212;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104243985A (en) | 2014-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104054344B (en) | Chroma data is deblocked for video coding | |
| JP6316487B2 (en) | Encoder, decoder, method, and program | |
| US20210144410A1 (en) | Method for encoding and decoding image information | |
| CN108184129B (en) | Video coding and decoding method and device and neural network for image filtering | |
| CN104041035B (en) | Lossless coding and coherent signal method for expressing for composite video | |
| CN105898335B (en) | Promote the loop circuit filtering method and loop filter of hardware efficiency | |
| KR102492116B1 (en) | Method and Apparatus for Reconciling Multiple Sign Bit Hiding and Residual Sign Prediction | |
| CN104303501B (en) | Quantization matrix and deblocking filter for video coding | |
| CN105532007B (en) | It is filtered in video coding for the adaptive de-blocking of non-luminance component | |
| CN103609124B (en) | Mode dependent intra smoothing filter table mapping methods for non-square prediction units | |
| CN104205836B (en) | The method and product for offseting and deblocking for colourity slicing layer level quantized parameter in Video coding | |
| KR20220140679A (en) | A method and an apparatus for decoding a video signal | |
| CN108605143A (en) | The method and apparatus of non local adaptive in-loop filter in Video coding | |
| CN111819853A (en) | Signaling residual symbols for prediction in transform domain | |
| CN109076237A (en) | The method and apparatus of the intra prediction mode of intra-frame prediction filtering device are used in video and compression of images | |
| SG191845A1 (en) | Moving image encoding device, moving image decoding device, moving image encoding method, and moving image decoding method | |
| CN110024405A (en) | Image processing method and its device | |
| CN107172424A (en) | Loop circuit filtering method and its equipment | |
| CN104243985B (en) | Sample adaptive equalization in HEVC | |
| CN105847842A (en) | Image encoding method and apparatus and image decoding method and apparatus | |
| WO2020096510A1 (en) | Asymmetric deblocking in a video encoder and/or video decoder | |
| AU2018415347B2 (en) | An image processing device and method for performing efficient deblocking | |
| EP3711302A1 (en) | Spatially adaptive quantization-aware deblocking filter | |
| CN107040778A (en) | Loop circuit filtering method and loop filter | |
| CN103442229A (en) | Bit rate estimation method of SAO mode decision applied to encoder of HEVC standard |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201104 Address after: Group 1, Shengli Ju, Jiang'an Town, Rugao City, Nantong City, Jiangsu Province, 226500 Patentee after: RUGAO ANRUI ELECTRONIC Co.,Ltd. Address before: 100000 Beijing Haidian District Huayuan Road No. 2 peony building 4 floor 1424 room Patentee before: Beijing Zhonglian Technology Service Co.,Ltd. Effective date of registration: 20201104 Address after: 100000 Beijing Haidian District Huayuan Road No. 2 peony building 4 floor 1424 room Patentee after: Beijing Zhonglian Technology Service Co.,Ltd. Address before: 100144 Beijing City, Shijingshan District Jin Yuan Zhuang Road No. 5, North China University of Technology Patentee before: North China University of Technology |