CN106375762B - Reference frame data compression method and its device - Google Patents
Reference frame data compression method and its device Download PDFInfo
- Publication number
- CN106375762B CN106375762B CN201510435316.8A CN201510435316A CN106375762B CN 106375762 B CN106375762 B CN 106375762B CN 201510435316 A CN201510435316 A CN 201510435316A CN 106375762 B CN106375762 B CN 106375762B
- Authority
- CN
- China
- Prior art keywords
- group
- data block
- pixel
- prediction mode
- 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.)
- Active
Links
Abstract
The present invention relates to field of image processing, a kind of reference frame data compression method and its device are disclosed.Reference frame the following steps are included: is divided into the data block of multiple predefined sizes by compression method of the invention；To each data block, the compression ratio when prediction mode using various candidates is calculated separately, and the highest prediction mode of compression ratio is selected compress to the data block；Wherein, the prediction mode of each candidate only uses each pixel in notebook data block and carries out block interior prediction, data block is divided into multiple groups respectively by the prediction mode of each candidate, each group includes multiple pixels, includes the prediction mode that pixel presses 45 degree or 135 degree directional spredings in group in the prediction mode of each candidate.The prediction mode that the present invention can select corresponding highest compression ratio individually compresses data block, to effectively improve the compression ratio and compression success rate of reference frame.
Description
Technical field
The present invention relates to field of image processings, in particular to reference frame data compress technique.
Background technique
With the continuous development of video encoding standard, all kinds of coding standards, such as a kind of H.264 (Video Codec mark
It is quasi-), H.265 (a kind of video encoding standard), VP9 (a kind of video compression standard) etc., the encoded picture size supported is more next
Bigger, high definition coding, 4K coding have become homely food.Becoming larger for dimension of picture directly results in the significantly increasing of reference frame data
Add, the readwrite bandwidth of encoder chip is also followed by increase, and in the encoder system of some bandwidth anxieties, readwrite bandwidth is to volume
The performance of code device will have a direct impact.
Currently, reference frame compression method can be simply divided into two classes: one kind is lossy compression, and one kind is lossless compression.
Relative to lossless compression, lossy compression has better compression ratio, but this can reduce the quality of coded image, thus most of volumes
Code device using lossless compression or damages the method combined with lossless compression.But current compression method is still deposited
The problems such as compression ratio is low, coding is not flexible, transport complexity is high, and coding is complicated, a kind of reference that compression ratio is high is developed thus
Frame compression method is most important for encoder.
Summary of the invention
The purpose of the present invention is to provide a kind of reference frame data compression method and its devices, can select corresponding maximal pressure
The prediction mode of shrinkage individually compresses data block, to effectively improve the compression ratio and compression success rate of reference frame.
In order to solve the above technical problems, embodiments of the present invention disclose a kind of reference frame data compression method, including
Following steps:
Reference frame is divided into the data block of multiple predefined sizes；
To each data block, the compression ratio when prediction mode using various candidates is calculated separately, and select compression ratio
Highest prediction mode compress to the data block；Wherein, the prediction mode of each candidate only uses in notebook data block
Each pixel carries out block interior prediction, and data block is divided into multiple groups respectively by the prediction mode of each candidate, and each group includes multiple
Pixel includes the prediction mode that pixel presses 45 degree or 135 degree directional spredings in group in the prediction mode of each candidate.
Embodiments of the present invention also disclose a kind of reference frame data compression set, comprising:
Cutting unit, for reference frame to be divided into the data block of multiple predefined sizes；
Compression unit, for calculating separately the compression ratio when prediction mode using various candidates to each data block,
And the highest prediction mode of compression ratio is selected compress to the data block；
Wherein, the prediction mode of each candidate only uses each pixel in notebook data block and carries out block interior prediction, each candidate's
Data block is divided into multiple groups respectively by prediction mode, and each group includes multiple pixels, includes in the prediction mode of each candidate
Pixel presses the prediction mode of 45 degree or 135 degree directional spredings in group.
Compared with prior art, the main distinction and its effect are embodiment of the present invention:
By individually calculating compression ratio of each data block of reference frame under different prediction modes, corresponding highest can select
The prediction mode of compression ratio individually compresses data block, to effectively improve the compression ratio and compression success rate of reference frame.
Further, the arithmetic path of decompression can be reduced, to reduce the pressure of decompression engine, with the data of 8x8
For block, the adder by 14 grades is needed from the initial predicted pixel that coordinate is (0,0) to the pixel that coordinate is (7,7), it is right
Compression engine it is more demanding, and when the coordinate of initial prediction pixel is changed to (3,3)/(3,4)/(4,3)/(4,4), longest path
Diameter from (3,3) to (7,7) need to only pass through 8 grades of adders, this is to optimization circuit sequence and arranges flowing water all very helpful.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of reference frame data compression method in first embodiment of the invention；
Fig. 2 a and 2b are a kind of grouping schematic diagrames of data block under prediction mode in first embodiment of the invention；
Fig. 3 a and 3b are a kind of grouping schematic diagrames of data block under prediction mode in first embodiment of the invention；
Fig. 4 a and 4b are a kind of grouping schematic diagrames of data block under prediction mode in first embodiment of the invention；
Fig. 5 a and 5b are a kind of grouping schematic diagrames of data block under prediction mode in first embodiment of the invention；
Fig. 6 is a kind of structural schematic diagram of reference frame data compression set in third embodiment of the invention.
Specific embodiment
In the following description, in order to make the reader understand this application better, many technical details are proposed.But this
The those of ordinary skill in field is appreciated that even if without these technical details and many variations based on the following respective embodiments
And modification, each claim of the application technical solution claimed can also be realized.
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to implementation of the invention
Mode is described in further detail.
First embodiment of the invention is related to a kind of reference frame data compression method.Fig. 1 is the reference frame data compression side
The flow diagram of method.
Specifically, as shown in Figure 1, the reference frame data compression method the following steps are included:
In a step 101, reference frame is divided into the data block of multiple predefined sizes.
Then into step 102, to each data block, the compression when prediction mode using various candidates is calculated separately
Rate, and the highest prediction mode of compression ratio is selected compress to the data block.Wherein, the prediction mode of each candidate only makes
Block interior prediction is carried out with each pixel in notebook data block, data block is divided into multiple groups respectively by the prediction mode of each candidate,
Each group includes multiple pixels, includes the interior pixel of group in the prediction mode of each candidate by 45 degree or 135 degree directional spredings
Prediction mode.
Hereafter, terminate this process.
In addition, in a preferred example, in a step 102, data block is predicted multiple groups that mode is divided and includes:
Pixel in one group of first kind group and at least two groups group being made of the pixel on 135 degree of diagonal lines is along 45 degree of directions
It is distributed in the second class group of 135 degree of diagonal line two sides, and the number of pixels in first kind group is number of pixels in the second class group
Half.Or the pixel in one group of first kind group and at least two groups group being made of the pixel on 45 degree of diagonal lines is along 135
Directional spreding is spent in the second class group of 45 degree of diagonal line two sides, and the number of pixels in first kind group is pixel number in the second class group
Purpose half.
Specifically, as shown in figures 2 a and 2b, by taking 8 × 8 data block as an example, the first kind group of the data block shown in Fig. 2 a by
Pixel composition on 135 degree of diagonal lines, wherein initial predicted pixel (black) is the 4th pixel on diagonal line, by the first kind
Group is divided into two first kind subgroups, and four group of second class group is divided by the pixel on 135 degree of diagonal lines respectively along 45 degree of directional spredings
Totally 8 the second class subgroups (it should be noted that in attached drawing of the present invention, the identical pixel of color forms one group).Fig. 2 b is shown
The first kind group of data block be made of the pixel on 45 degree of diagonal lines, wherein initial predicted pixel is the 4th on diagonal line
First kind group is divided into two first kind subgroups by pixel, and four group of second class group is along 135 degree of directional spredings, respectively by 45 degree pairs
Pixel on linea angulata is divided into totally 8 the second class subgroups.
In addition, further including the first rotation prediction mode, data block quilt in the prediction mode of each candidate in another preferred example
First, which rotates multiple groups that prediction mode is divided, includes:
One group of first kind group being made of the pixel in a line in two row of data block center and at least two groups are by the number
The the second class group formed according to the adjacent two column pixel of block.Wherein, the initial predicted pixel of the prediction mode is located at first kind group institute
Row, first kind group is divided into two first kind subgroups by initial predicted pixel, the prediction directions of two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the row where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between subgroup changes clockwise or counterclockwise.Specifically, as schemed
Shown in 3a, by taking 8 × 8 data block as an example, the first kind group of the data block shown in Fig. 3 a is made of 7 pixels in fourth line,
Initial predicted pixel coordinate is (4,4), i.e. fourth line the 4th arranges, and first kind group is divided into two groups of first kind subgroups, wherein organizing a1
Prediction direction to the left, a2Prediction direction to the right.Data block adjacent column forms the second class group, wherein the 1st column and the 2nd column composition
The second class group be divided into b1And b2Two the second class subgroups, prediction direction be respectively upwards and to the left, and so on, prediction side
To with a1→b1+c1→d1+e1→a2→d2+e2→c2+b2Sequence rotate clockwise variation.
Furthermore, it is to be understood that in other embodiments of the invention, by the data after the first rotation prediction mode grouping
In block the prediction direction of each group can also with the difference in Fig. 3 a, for example, prediction direction is with a1→b1+c1→d1→e1→a2→d2
→e2→c2→b2Sequence along clockwise or counterclockwise variation, etc..
It include the second rotation prediction mode in addition, in another preferred example, in the prediction mode of each candidate, data block is by the
Two, which rotate multiple groups that prediction modes are divided, includes:
One group of first kind group being made of the pixel on the column in the column of the data block center two and at least two groups are by the number
The the second class group formed according to the adjacent rows pixel of block.Wherein, the initial predicted pixel of the prediction mode is located at first kind group institute
Column, first kind group is divided into two first kind subgroups by initial predicted pixel, the prediction directions of two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the column where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between subgroup changes clockwise or counterclockwise.Specifically, as schemed
Similar with Fig. 3 a by taking 8 × 8 data block as an example shown in 3b, difference is that first kind group is made of the pixel of the 4th column, second
Class group is made of adjacent rows.Wherein, prediction direction is with a3→b3+c3→d3→e3→a4→d4+e4→c4+b4Sequence along the inverse time
Needle is rotationally-varying.It similarly, can also be with Fig. 3 b by the prediction direction of each group in the data block after the first rotation prediction mode grouping
In difference.
Furthermore, it is to be understood that the diversity based on image texture variation, candidate in other embodiments of the invention
Prediction mode can also include other prediction modes, such as horizontal prediction mode (as shown in figures 4 a and 4b), vertical prediction mode
(as shown in figure 5 a and 5b) etc..Wherein, prediction direction is as shown by arrows, indicates that the horizontal line expression in the grid of pixel takes phase
The average value of adjacent two pixels, takes the average value of the two neighboring pixel of same row in fig. 4b, takes same a line adjacent in figure 5b
The average value of two pixels.
Preferably, the data block in the application includes NxN pixel, and wherein N is the positive integer greater than 4.Each candidate's is pre-
The initial predicted pixel of survey mode is one in the pixel at data block center.
Using the pixel at center as initial predicted pixel, the arithmetic path of decompression can be reduced, is drawn to reduce decompression
The pressure held up, as shown in Figs. 4a and 4b, by taking the data block of 8x8 as an example, from coordinate be (0,0) initial predicted pixel to coordinate be
The pixel of (7,7) needs the adder by 14 grades, to the more demanding of compression engine, and when the coordinate of initial prediction pixel changes
When for (3,3)/(3,4)/(4,3)/(4,4), longest path from (3,3) to (7,7) need to only pass through 8 grades of adders, this is to optimization
Circuit sequence and arrangement flowing water are all very helpful.
Furthermore, it is to be understood that center pixel can not also be used as initial pre- in the other embodiments of the application
Pixel is surveyed, herein with no restrictions.
In another preferred example, above-mentioned steps 102 include following sub-step:
The difference in data block between the predicted value and true value of each pixel is calculated according to current prediction mode.According to number
The difference of maximum absolute value determines the coding mode of respective sets pixel in each group be divided into according to block.It calculates true using each group
Fixed corresponding encoded mode carries out each group to encode required bit number, and determines that data block exists based on resulting bit number is calculated
Compression ratio under current prediction mode.
In addition, the difference of maximum absolute value determines the volume of respective sets pixel in each group be divided into according to data block
In the sub-step of pattern, if maximum value in absolute difference is less than or equal to 64, to group corresponding to the maximum value into
When row coding, the complement of two's two's complement digit of the coding mode of use is less than or equal to 7, and when maximum value is 0 or 1, use
The complement of two's two's complement digit of coding mode is 1.
According to the characteristic of reference frame, by individually calculating compression of each data block of reference frame under different prediction modes
Rate, the prediction mode that can select corresponding highest compression ratio individually compresses data block, to effectively improve reference frame
Compression ratio and compression success rate.
Second embodiment of the invention is related to a kind of reference frame data compression method.In this embodiment, with 8 × 8
For data block, specifically, comprising the following steps:
First, using one of candidate modes described in first embodiment, in addition to Center Prediction pixel
Each pixel calculate predicted value.
Second, the true value of each pixel is subtracted into predicted value and obtains the difference of each pixel.
63 pixels in addition to Center Prediction pixel are divided into one first according to current prediction mode by third
Class group and 4 the second class groups, wherein first kind group is divided into two first kind subgroups, Mei Ge using Center Prediction pixel as boundary
Two class groups are divided into four the second class subgroups by first kind group.
4th, calculate the coding mode for each subgroup that third step obtains.Specifically, it is determined that maximum difference is exhausted in subgroup
To the data area that mm is arranged in affiliated table 1 is worth, coding mode (code mode, CM) corresponding to determining data area is i.e.
For the coding mode of the subgroup.Wherein, in table 1, the maximum value in absolute difference is less than or equal to 64, then to the maximum value institute
Corresponding group when being encoded, the complement of two's two's complement digit of the coding mode of use is less than or equal to 7, and when maximum value is 0 or 1
When, the complement of two's two's complement digit of the coding mode used is 1.S is sign bit.
1. coding schedule of table
It is appreciated that other coding modes can also be used to determine for each subgroup in other embodiments of the invention
Coding mode.
5th, it is merged according to coding mode of such as under type to each subgroup: firstly, by the identical subgroup of coding mode
It merges, the unique subgroup of coding mode does not do merging treatment；Secondly, if there are pixels in first kind group and the second class group
The subgroup that number is 3, and there are one bigger than the coding mode of the subgroup 1 subgroup, then the two subgroups are merged, and be
Group after merging uses biggish coding mode.
6th, combination situation is obtained according to the 5th step, calculates and bit number (packet required for encoding respectively is completed according to table 1
Include the mark of necessary maximum value symbol when coding mode, difference data and the table 1 of all subgroups (including merging subgroup) is tabled look-up)
(first kind group only has 2 kinds of reform patterns to bit number required for encoding with reform patterns, and bit is needed to indicate, the second class group has
15 kinds of reform patterns, if the second class group merges the group being made of at one two subgroups, with 3 bits, 4 ratios of other situations
It is special)；
7th, the bit number after calculating other candidate modes compression data blocks by the way of the first to the 6th step,
Select the highest prediction mode of compression ratio as the final prediction mode of 8X8 data block.
8th, reference frame is compressed using final prediction mode is selected, calculates compressed bit number, including in
Heart predicted pixel values (8 bit), prediction mode (3 bit), the reform patterns (13~17 bit) of 8X8 data block, first kind group
With the coding modes (15~54 bit) of all second class groups, all differential codings and table look-up 1 when necessary maximum value symbol
Mark, this block is the motionless block of pressure if bit number is greater than 512 after compression, uses former data block as final compression result, otherwise
Data are organized in order as final compression result in the order described above.
In addition, being tested using the above method 8X8 data block compression result, test has used 8 different 720P
Sequence, compared with existing compression method, compression ratio (CR) that the present invention obtains, compression are more than half ratio (≤256) and press
The case where contracting Success Ratio (≤512), is as shown in table 2:
2 compression result deck watch of table
Wherein, Qp indicates quantization parameter.From table 2 it can be seen that compared with HAC+SBT (a kind of existing compression method),
With the increase of Qp, (the compressed bit number of data block is divided by 512 institutes for a kind of MDA+SFL (existing compression method) compression ratio
, it is lower to illustrate that compression ratio is higher) and compression be more than half (the higher the better) advantage be more and more obvious, although compression successfully
Disadvantage is slowed down, but is calculated according to the breadth of 720P, and compression failure number still differs very more；Compared with HAC+SBT,
The institute that the present invention maintains MDA_SFL substantially is advantageous, and it is very close with HAC+SBT to compress successful number；With MDA+SFL phase
Than, when QP is less than normal, compression ratio of the invention and compress it is successfully advantageous, when QP is bigger than normal, although compression ratio is relatively poor,
But the successful advantage of compression still remains.
Each method embodiment of the invention can be realized in a manner of software, hardware, firmware etc..Regardless of the present invention be with
Software, hardware or firmware mode realize that instruction code may be stored in any kind of computer-accessible memory
In (such as permanent perhaps revisable volatibility is perhaps non-volatile solid or non-solid, it is fixed or
The replaceable medium etc. of person).Equally, memory may, for example, be programmable logic array (Programmable Array
Logic, referred to as " PAL "), random access memory (Random Access Memory, referred to as " RAM "), it may be programmed read-only deposit
Reservoir (Programmable Read Only Memory, referred to as " PROM "), read-only memory (Read-Only Memory, letter
Claim " ROM "), electrically erasable programmable read-only memory (Electrically Erasable Programmable ROM, referred to as
" EEPROM "), disk, CD, digital versatile disc (Digital Versatile Disc, referred to as " DVD ") etc..
Third embodiment of the invention is related to a kind of reference frame data compression set.Fig. 6 is reference frame data compression dress
The structural schematic diagram set.
Specifically, the reference frame data compression set includes:
Cutting unit, for reference frame to be divided into the data block of multiple predefined sizes.
Compression unit, for calculating separately the compression ratio when prediction mode using various candidates to each data block,
And the highest prediction mode of compression ratio is selected compress to the data block.Wherein, the prediction mode of each candidate only uses
Each pixel in notebook data block carries out block interior prediction, and data block is divided into multiple groups respectively by the prediction mode of each candidate, often
A group includes multiple pixels, includes that the interior pixel of group presses the pre- of 45 degree or 135 degree directional spredings in the prediction mode of each candidate
Survey mode.
In addition, in a preferred example, data block is organized interior pixel by the prediction mode point of 45 degree or 135 degree directional spredings
Multiple groups cut include:
Pixel in one group of first kind group and at least two groups group being made of the pixel on 135 degree of diagonal lines is along 45 degree of directions
It is distributed in the second class group of 135 degree of diagonal line two sides, and the number of pixels in first kind group is number of pixels in the second class group
Half.Or the pixel in one group of first kind group and at least two groups group being made of the pixel on 45 degree of diagonal lines is along 135
Directional spreding is spent in the second class group of 45 degree of diagonal line two sides, and the number of pixels in first kind group is pixel number in the second class group
Purpose half.
It in another preferred example, include the first rotation prediction mode in the prediction mode of each candidate, data block is by the first rotation
Turning multiple groups that prediction mode is divided includes:
One group of first kind group being made of the pixel in a line in two row of data block center and at least two groups are by the number
The the second class group formed according to the adjacent two column pixel of block.Wherein, the initial predicted pixel of the prediction mode is located at first kind group institute
Row, first kind group is divided into two first kind subgroups by initial predicted pixel, the prediction directions of two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the row where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between subgroup changes clockwise or counterclockwise.
It in another preferred example, include the second rotation prediction mode in the prediction mode of each candidate, data block is by the second rotation
Turning multiple groups that prediction mode is divided includes:
One group of first kind group being made of the pixel on the column in the column of the data block center two and at least two groups are by the number
The the second class group formed according to the adjacent rows pixel of block.Wherein, the initial predicted pixel of the prediction mode is located at first kind group institute
Column, first kind group is divided into two first kind subgroups by initial predicted pixel, the prediction directions of two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the column where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between subgroup changes clockwise or counterclockwise.
In addition, in another preferred example, above-mentioned compression unit includes following subelement:
Computation subunit, for according to current prediction mode calculate data block in each pixel predicted value and true value it
Between difference.
Coded sub-units, the difference of maximum absolute value determines respective sets in each group for being divided into according to data block
The coding mode of pixel.And coded sub-units are when determining coding mode, if the maximum value in absolute difference is less than etc.
In 64, then when encoding to group corresponding to the maximum value, the complement of two's two's complement digit of the coding mode of use is less than or equal to
7, and when maximum value is 0 or 1, the complement of two's two's complement digit of the coding mode used is 1.
Subelement is compressed, each group is carried out to encode required bit for calculating the corresponding encoded mode for using each group to determine
Number, and compression ratio of the data block under current prediction mode is determined based on resulting bit number is calculated.
In addition, in another preferred example, data block includes NxN pixel, wherein N is the positive integer greater than 4.Each candidate's
The initial predicted pixel of prediction mode is one in the pixel at data block center.
First embodiment is method implementation corresponding with present embodiment, and present embodiment can be implemented with first
Mode is worked in coordination implementation.The relevant technical details mentioned in first embodiment are still effective in the present embodiment, in order to
It reduces and repeats, which is not described herein again.Correspondingly, the relevant technical details mentioned in present embodiment are also applicable in the first implementation
In mode.
It should be noted that each unit mentioned in each equipment embodiment of the present invention is all logic unit, physically,
One logic unit can be a physical unit, be also possible to a part of a physical unit, can also be with multiple physics
The combination of unit realizes that the Physical realization of these logic units itself is not most important, these logic units institute reality
The combination of existing function is only the key for solving technical problem proposed by the invention.In addition, in order to protrude innovation of the invention
Part, there is no the technical problem relationship proposed by the invention with solution is less close for the above-mentioned each equipment embodiment of the present invention
Unit introduce, this does not indicate above equipment embodiment and there is no other units.
It should be noted that in the claim and specification of this patent, such as first and second or the like relationship
Term is only used to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying
There are any actual relationship or orders between these entities or operation.Moreover, the terms "include", "comprise" or its
Any other variant is intended to non-exclusive inclusion so that include the process, methods of a series of elements, article or
Equipment not only includes those elements, but also including other elements that are not explicitly listed, or further include for this process,
Method, article or the intrinsic element of equipment.In the absence of more restrictions, being wanted by what sentence " including one " limited
Element, it is not excluded that there is also other identical elements in the process, method, article or apparatus that includes the element.
Although being shown and described to the present invention by referring to some of the preferred embodiment of the invention,
It will be understood by those skilled in the art that can to it, various changes can be made in the form and details, without departing from this hair
Bright spirit and scope.
Claims (11)
1. a kind of reference frame data compression method, which comprises the following steps:
Reference frame is divided into the data block of multiple predefined sizes；
To each data block, the compression ratio when prediction mode using various candidates is calculated separately, and select compression ratio highest
Prediction mode the data block is compressed；Wherein, the prediction mode of each candidate only uses each pixel in notebook data block
Block interior prediction is carried out, data block is divided into multiple groups respectively by the prediction mode of each candidate, and each group includes multiple pixels, respectively
It include the prediction mode that pixel presses 45 degree or 135 degree directional spredings in group in candidate prediction mode；
The step of " to each data block, compression ratio " when calculating separately the prediction mode using various candidates include with
Lower sub-step:
The difference in data block between the predicted value and true value of each pixel is calculated according to current prediction mode；
The difference of maximum absolute value determines the coding mode of respective sets pixel in each group be divided into according to data block；
It calculates the corresponding encoded mode determined using each group each group is carried out to encode required bit number, and resulting based on calculating
The bit number determines compression ratio of the data block under current prediction mode.
2. reference frame data compression method according to claim 1, which is characterized in that the data block is by the prediction mould
Formula segmentation multiple groups include:
Pixel in one group of first kind group and at least two groups group being made of the pixel on 135 degree of diagonal lines is along 45 degree of directional spredings
The second class group in 135 degree of diagonal line two sides, and the number of pixels in first kind group is two points of number of pixels in the second class group
One of；Or
Pixel in one group of first kind group and at least two groups group being made of the pixel on 45 degree of diagonal lines is along 135 degree of directional spredings
The second class group in 45 degree of diagonal line two sides, and the number of pixels in first kind group be the second class group in number of pixels two/
One.
3. reference frame data compression method according to claim 1, which is characterized in that include in the prediction mode of each candidate
First rotation prediction mode, the data block include: by multiple groups of the first rotation prediction mode segmentation
One group of first kind group being made of the pixel in a line in two row of data block center and at least two groups are by the data block
Adjacent two column pixel composition the second class group；Wherein, the initial predicted pixel of the prediction mode is located at where first kind group
Row, first kind group are divided into two first kind subgroups by initial predicted pixel, the prediction directions of described two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the row where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between the subgroup changes clockwise or counterclockwise.
4. reference frame data compression method according to claim 1, which is characterized in that include in the prediction mode of each candidate
Second rotation prediction mode, the data block include: by multiple groups of the second rotation prediction mode segmentation
One group of first kind group being made of the pixel on the column in the column of the data block center two and at least two groups are by the data block
Adjacent rows pixel composition the second class group；Wherein, the initial predicted pixel of the prediction mode is located at where first kind group
Column, first kind group are divided into two first kind subgroups by initial predicted pixel, the prediction directions of described two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the column where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between the subgroup changes clockwise or counterclockwise.
5. reference frame data compression method according to claim 1, which is characterized in that be divided described according to data block
At each group in the difference of maximum absolute value determine in the sub-step of coding mode of respective sets pixel, if difference is exhausted
64, then when encoding to group corresponding to the maximum value, the two of the coding mode of use are less than or equal to the maximum value in value
Complement on n n digit is less than or equal to 7, and when the maximum value is 0 or 1, the complement of two's two's complement digit of the coding mode of use
It is 1.
6. reference frame data compression method according to any one of claim 1 to 5, which is characterized in that the data block
Including NxN pixel, wherein N is the positive integer greater than 4；
The initial predicted pixel of the prediction mode of each candidate is one in the pixel at the data block center.
7. a kind of reference frame data compression set characterized by comprising
Cutting unit, for reference frame to be divided into the data block of multiple predefined sizes；
Compression unit, for calculating separately the compression ratio when prediction mode using various candidates, and select to each data block
It selects compression ratio highest prediction mode the data block compress；
Wherein, the prediction mode of each candidate only uses each pixel in notebook data block and carries out block interior prediction, the prediction of each candidate
Data block is divided into multiple groups respectively by mode, and each group includes multiple pixels, includes in group in the prediction mode of each candidate
Pixel presses the prediction mode of 45 degree or 135 degree directional spredings；
The compression unit includes following subelement:
Computation subunit, for being calculated in data block between the predicted value and true value of each pixel according to current prediction mode
Difference；
Coded sub-units, the difference of maximum absolute value determines respective sets in each group for being divided into according to data block
The coding mode of pixel；
Subelement is compressed, each group is carried out to encode required bit number for calculating the corresponding encoded mode for using each group to determine,
And compression ratio of the data block under current prediction mode is determined based on the resulting bit number is calculated.
8. reference frame data compression set according to claim 7, which is characterized in that include in the prediction mode of each candidate
First rotation prediction mode, the data block include: by multiple groups of the first rotation prediction mode segmentation
One group of first kind group being made of the pixel in a line in two row of data block center and at least two groups are by the data block
Adjacent two column pixel composition the second class group；Wherein, the initial predicted pixel of the prediction mode is located at where first kind group
Row, first kind group are divided into two first kind subgroups by initial predicted pixel, the prediction directions of described two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the row where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between the subgroup changes clockwise or counterclockwise.
9. reference frame data compression set according to claim 7, which is characterized in that include in the prediction mode of each candidate
Second rotation prediction mode, the data block include: by multiple groups of the second rotation prediction mode segmentation
One group of first kind group being made of the pixel on the column in the column of the data block center two and at least two groups are by the data block
Adjacent rows pixel composition the second class group；Wherein, the initial predicted pixel of the prediction mode is located at where first kind group
Column, first kind group are divided into two first kind subgroups by initial predicted pixel, the prediction directions of described two first kind subgroups on the contrary,
Each second class group is divided into two the second class subgroups by the column where first kind group, belongs to two second classes of a second class group
The prediction direction of subgroup is mutually perpendicular to, and the prediction direction edge between the subgroup changes clockwise or counterclockwise.
10. reference frame data compression set according to claim 7, which is characterized in that
The coded sub-units are when determining coding mode, if the maximum value in the absolute difference is less than or equal to 64,
When encoding to group corresponding to the maximum value, the complement of two's two's complement digit of the coding mode of use is less than or equal to 7, and works as
When the maximum value is 0 or 1, the complement of two's two's complement digit of the coding mode used is 1.
11. reference frame data compression set according to any one of claims 7 to 10, which is characterized in that the data
Block includes NxN pixel, and wherein N is the positive integer greater than 4；
The initial predicted pixel of the prediction mode of each candidate is one in the pixel at the data block center.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510435316.8A CN106375762B (en) | 2015-07-22 | 2015-07-22 | Reference frame data compression method and its device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510435316.8A CN106375762B (en) | 2015-07-22 | 2015-07-22 | Reference frame data compression method and its device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106375762A CN106375762A (en) | 2017-02-01 |
| CN106375762B true CN106375762B (en) | 2019-05-24 |
Family
ID=57880155
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510435316.8A Active CN106375762B (en) | 2015-07-22 | 2015-07-22 | Reference frame data compression method and its device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106375762B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108810556B (en) * | 2017-04-28 | 2021-12-24 | 炬芯科技股份有限公司 | Method, device and chip for compressing reference frame |
| CN107582046B (en) * | 2017-09-18 | 2022-03-01 | 山东正心医疗科技有限公司 | Real-time electrocardio monitoring method |
| CN110933444B (en) * | 2018-09-20 | 2021-10-29 | 北京君正集成电路股份有限公司 | Bit width value storage method and device |
| CN110933413A (en) * | 2018-09-20 | 2020-03-27 | 北京君正集成电路股份有限公司 | Video frame processing method and device |
| CN110007855B (en) * | 2019-02-28 | 2020-04-28 | 华中科技大学 | Hardware-supported 3D stacked NVM (non-volatile memory) memory data compression method and system |
| CN110175185B (en) * | 2019-04-17 | 2023-04-07 | 上海天数智芯半导体有限公司 | Self-adaptive lossless compression method based on time sequence data distribution characteristics |
| CN111866519A (en) * | 2019-04-30 | 2020-10-30 | 上海富瀚微电子股份有限公司 | Coding method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101431671A (en) * | 2007-11-07 | 2009-05-13 | 财团法人工业技术研究院 | Methods for selecting a prediction mode and encoder thereof |
| CN103125118A (en) * | 2010-10-01 | 2013-05-29 | 索尼公司 | Image processing device and image processing method |
| CN103248891A (en) * | 2013-04-24 | 2013-08-14 | 复旦大学 | N-bit truncation quantization and in-block two-dimensional prediction-based reference frame compression method |
| CN103563389A (en) * | 2011-06-03 | 2014-02-05 | 高通股份有限公司 | Intra prediction mode coding with directional partitions |
| CN103988437A (en) * | 2011-11-08 | 2014-08-13 | 高通股份有限公司 | Context reduction for context adaptive binary arithmetic coding |
| CN103997646A (en) * | 2014-05-13 | 2014-08-20 | 北京航空航天大学 | Rapid intra-frame prediction mode selection method in high-definition video coding |
-
2015
- 2015-07-22 CN CN201510435316.8A patent/CN106375762B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101431671A (en) * | 2007-11-07 | 2009-05-13 | 财团法人工业技术研究院 | Methods for selecting a prediction mode and encoder thereof |
| CN103125118A (en) * | 2010-10-01 | 2013-05-29 | 索尼公司 | Image processing device and image processing method |
| CN103563389A (en) * | 2011-06-03 | 2014-02-05 | 高通股份有限公司 | Intra prediction mode coding with directional partitions |
| CN103988437A (en) * | 2011-11-08 | 2014-08-13 | 高通股份有限公司 | Context reduction for context adaptive binary arithmetic coding |
| CN103248891A (en) * | 2013-04-24 | 2013-08-14 | 复旦大学 | N-bit truncation quantization and in-block two-dimensional prediction-based reference frame compression method |
| CN103997646A (en) * | 2014-05-13 | 2014-08-20 | 北京航空航天大学 | Rapid intra-frame prediction mode selection method in high-definition video coding |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106375762A (en) | 2017-02-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106375762B (en) | Reference frame data compression method and its device | |
| CN103237216B (en) | The decoding method and coding and decoding device of depth image | |
| CN103797800B (en) | It derives reference model value and coding and decoding represents the information of prediction mode | |
| TWI543083B (en) | Method of creating two dimensional code | |
| CN103067715A (en) | Encoding and decoding methods and encoding and decoding device of range image | |
| CN106412579B (en) | A kind of coding of image, coding/decoding method and device | |
| US9967565B2 (en) | Image compression method based on local dynamic quantizing | |
| CN113784124B (en) | Block matching encoding and decoding method for fine division using multi-shape sub-blocks | |
| CN101945273A (en) | Compression-encoding device and Visual Display control device | |
| CN102611880A (en) | Encoding method and device for marking geometric classification mode of image block | |
| CN103931181A (en) | Adaptive upsampling for spatially scalable video coding | |
| US20180184096A1 (en) | Method and apparatus for encoding and decoding lists of pixels | |
| JP5652101B2 (en) | Image processing apparatus and image processing method | |
| Zhao et al. | Three-dimensional histogram shifting for reversible data hiding | |
| CN111432212A (en) | Intra-frame division method and system based on texture features and storage medium | |
| WO2013009224A1 (en) | Multi-mode processing of texture blocks | |
| US9020283B2 (en) | Electronic device and method for splitting image | |
| CN102542528B (en) | Image conversion processing method and system | |
| CN114079792A (en) | Compression method, coding method and device for video coding reference block | |
| CN107103632B (en) | Image compression method and device | |
| CN110213595B (en) | Intra-frame prediction based encoding method, image processing apparatus, and storage device | |
| CN107454425A (en) | A kind of SCC intraframe codings unit candidate modes reduction method | |
| US8675975B2 (en) | Method for encoding image using estimation of color space | |
| CN113170134A (en) | Block-based picture fusion for context segmentation and processing | |
| CN113365080B (en) | Encoding and decoding method, device and storage medium for string coding technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |