CN1589020A - Fractional pixel filter system and its filter method for video frequency decoding chip - Google Patents
Fractional pixel filter system and its filter method for video frequency decoding chip Download PDFInfo
- Publication number
- CN1589020A CN1589020A CN 200410070363 CN200410070363A CN1589020A CN 1589020 A CN1589020 A CN 1589020A CN 200410070363 CN200410070363 CN 200410070363 CN 200410070363 A CN200410070363 A CN 200410070363A CN 1589020 A CN1589020 A CN 1589020A
- Authority
- CN
- China
- Prior art keywords
- filtering
- pixel
- decoding chip
- video decoding
- filter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims abstract description 38
- 230000004087 circulation Effects 0.000 claims abstract description 20
- 230000005540 biological transmission Effects 0.000 claims abstract description 13
- 238000001914 filtration Methods 0.000 claims description 83
- 230000000694 effects Effects 0.000 claims description 5
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 8
- 101150039623 Clip1 gene Proteins 0.000 description 7
- 101100060194 Caenorhabditis elegans clip-1 gene Proteins 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 241000288673 Chiroptera Species 0.000 description 3
- 230000008520 organization Effects 0.000 description 3
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Abstract
This invention relates to a video decoding chip fraction picture element filter system and its filter method including a memory, a data transmission circuit with a register array, a low-pass filter with four data selectors based on a pipeline. The data transmission circuit is connected with an external memory and stores the integral pixel matrix received from the external memory controller interface in a memory in the line mode, the register array circulation movers from the integral position delivered by the memory and sends the motion integral pixel to the low-pass filter for carrying out 1/2 and 1/4 filter. The data selector controls the data flowing mode to output fraction pixel position.
Description
Technical field
The invention belongs to digit image coding/decoding technology and SOC (system on a chip) technology (SOC, system onchip) technical field, specifically, is a kind of video decoding chip fraction pixel filtering system and filtering method thereof based on the AVS standard.
Background technology
The effective video encoding and decoding technique is to realize multi-medium data storage and the key of transmitting, and advanced video coding and decoding technology exists with the form of standard usually.The MPEG series international standard that the Motion Picture Experts Group (MPEG, MovingPicture Expert Group) that at present typical video compression standard has International Organization for Standardization to divide into releases; The H.26x series video compression standard that International Telecommunication Union proposes, and the video encoding standard A VS that formulating of China oneself.
Motion search is an of paramount importance part in the video encoding standard, traditional motion search all can only search integer-pel precision, but in order to improve search precision, new video standard such as mpeg4, h.264 all support the search of fraction pixel precision, AVS also supports the search of 1/4 pixel precision, in order to realize the search of fraction pixel precision, must use low pass filter to obtain the value of fraction pixel.But low pass filter algorithm is very complicated, and as if the time of realizing with software then needing to exhaust decoder nearly 1/3rd, decoding efficiency is lower.Simultaneously, the data multiplex rate is also lower, has wasted the bandwidth of memory.
Summary of the invention
At above-mentioned problems of the prior art, the purpose that the present invention carries is to provide a kind of video decoding chip fraction pixel filtering system and filtering method thereof, to improve the efficient that low pass filter carries out the fraction pixel precision search.
For finishing the foregoing invention purpose, the technical solution used in the present invention is: video decoding chip fraction pixel filtering system, comprise memory, the data transmission circuit of band register array, the low pass filter of band data selector, low pass filter is based on the low pass filter of streamline, data transmission circuit is connected with external memory, to deposit register array in the form of going from the whole picture element matrix that the external memory control unit interface receives, the whole location of pixels of being sent by memory is moved in the register array circulation, and the whole pixel of the activity after will circulating is sent into low pass filter, low pass filter carries out 1/2 and 1/4 filtering, data selector is exported the fraction pixel position according to the type of flow of the Position Control data of fraction pixel.
Described data selector is 4.
Described whole pixel needs the whole picture element matrix of one 13 * 13 brightness; Described whole pixel needs 5 * 5 the whole picture element matrix of colourity.
Described register array is divided into two parts, and wherein top unit is active unit all the time, is 6 * 13 matrix.
The whole pixel of described activity is one 6 * 6 a matrix.
Described low pass filter is 4 tap F
1(1,5,5 ,-1) filter is predicted 1/2 sample position of nearest pixel samples.
Described low pass filter is 4 tap F
2(1,7,7,1) filter is predicted 1/4 sample position of nearest pixel samples.
Video decoding chip fraction pixel filtering method may further comprise the steps at least:
The pixel of the whole picture element matrix of being sent by memory is moved in step 3, register array circulation, and will send into low pass filter through the pixel of the mobile movable integer pixel matrix that obtains of circulation;
Step 4, low pass filter carry out 1/2 and 1/4 filtering to the pixel of movable integer pixel matrix, obtain the fraction pixel position;
Step 5, the fraction pixel position is exported.
Register array circulation in the described step 3 is moved and be may further comprise the steps:
Step 31, in beginning 6 clock cycle, each cycle is taken out delegation's totally 13 pixels from memory, move into register array in the mode of moving horizontally;
Step 32, in 7 cycles next, array will be done in the circulation and move, it is effective to remain 6 * 6 top pixels;
Step 33, move into new one-row pixels with the inverted level of putting in order, 7 pixels of back are moved to the front in the 8th cycle;
Step 34, array are done circulation and are moved down;
Step 35,7 cycles move horizontally later again;
Whether the whole pixel of step 36, judgement input has moved, and intact then end is moved, otherwise repeating step 1.
Filtering in the described step 4 specifically may further comprise the steps:
Step 411, use F
1, F
24 integral sample filtering to nearest on level or the vertical direction obtain median;
Step 412, median are through after moving to right, and if resulting value is greater than 255, and then final predicted value is 255, and less than 0, then final predicted value is 0 as if the value that obtains, otherwise final predicted value is the resulting value that moves to right.
Described filtering specifically may further comprise the steps:
Step 421, use F
1, F
2Nearest 4 integers or fractional samples on the tilted direction are carried out 1/2 or 1/4 sample filtering, obtain median;
Step 422, median are through after moving to right, and if resulting value is greater than 255, and then final predicted value is 255, and less than 0, then final predicted value is 0 as if the value that obtains, otherwise final predicted value is the resulting value that moves to right.
Filtering in the described step 4 adopts multiplication to become the method for displacement and addition, divides 2-7 clock cycle to finish.
Filtering in the described step 4 is that the clock cycle is 2 with 1 1/2 filter filtering the 1st class point.
Filtering in the described step 4 is that the clock cycle is 4 with 2 1/2 filters and 1 1/4 filter filtering the 2nd class point.
Filtering in the described step 4 is that the clock cycle is 4 or 5 with 5 1/2 filters and 1 weighted average device filtering the 3rd class point.
Filtering in the described step 4 is that the clock cycle is 6 with 7 1/2 filters and 1 1/4 filter filtering the 4th class point.
The present invention has tangible advantage and good effect, based on video encoding standard, proposed a kind of based on streamline multiplexing filtering system and filtering method, realized the motion search of fraction pixel with hardware mode, solved with software mode and carried out problems such as decoder efficiency that the fraction pixel precision search time brought is low by low pass filter, save the decode time of Video Decoder, improved the decoding efficiency of Video Decoder.Be applicable to the interpolation of video decoding chip fraction precision pixel.
Description of drawings
Fig. 1 is a luminance component image sample matrix interpolation schematic diagram;
Fig. 2 is forecast sample motion vector figure;
Fig. 3 is a chroma sample interpolation schematic diagram;
Fig. 4 is fraction pixel filtering system modular structure figure of the present invention;
Fig. 5 is a register array cyclic shift schematic diagram;
Fig. 6 is luminance filter 1/2 a filtering schematic diagram;
Fig. 7 is to the 1st class point filter logical circuitry;
Fig. 8 is to the 2nd class point filter logical circuitry;
Fig. 9 is to the 3rd class point filter logical circuitry;
Figure 10 is to the 4th class point filter logical circuitry;
Figure 11 is a chrominance filter sample interpolation schematic diagram.
Embodiment
The present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.
See also Fig. 1, this figure has provided the position of reference picture luminance component integral sample, 1/2nd samples and 1/4th samples, what wherein use the capitalization mark is the integral sample position, and with the lowercase mark is 1/2nd and four/the same this position.
The predicted value of two/the same this position is by 4 tap filter F
1(1,5,5 ,-1) calculates.The predicted value of four/the same this position is by 4 tap filter F
2(1,7,7,1) calculate.
The computational process of 1/2nd samples is as follows:
/ 2nd sample b: at first use F
1To 4 nearest on horizontal direction integral sample filtering, obtain median b '=(C+5D+5E-F); Final predicted value b=Clip1 ((b '+4)＞＞3).
/ 2nd sample h: at first use F
1To 4 nearest on vertical direction integral sample filtering, obtain median h '=(A+5D+5H-K); Final predicted value h=Clip1 ((h '+4)＞＞3).
/ 2nd sample j: at first use F
1On level or vertical direction,, obtain median j '=(bb '+5h '+5m '-cc '), perhaps j '=(aa '+5b '+5s '-dd ') to nearest 4 1/2nd sample medians filtering.Aa ' wherein, dd ' and s ' they are relevant position 1/2nd sample medians, they use F
1Filtering in the horizontal direction obtains, and bb ', cc ' and m ' are relevant position 1/2nd sample medians, and they use F
1Obtain in vertical direction filtering.Final predicted value is j=Clip1 ((j '+32)＞＞6).The value that adopts horizontal direction or vertical direction filtering to obtain is identical.
Illustrate: top employed clip1 is a cutting function, and greater than 255, then output is 255 as if the number of importing, if import less than 0, then exporting is 0, equals input otherwise export.
The computational process of 1/4th samples is as follows:
/ 4th sample a: at first use F
2In the horizontal direction to ee ', D ', four value filterings of b ' and E ' obtain median a '=(ee '+7D '+7b '+E '); Final predicted value a=Clip1 ((a '+64)＞＞7).Wherein ee ' and b ' are relevant position 1/2nd sample medians, and D ' and E ' are that the relevant position integral sample is amplified 8 times value.
/ 4th sample d: at first with F2 in vertical direction to ff ', D ', four value filterings of h ' and H ' obtain median d '=(ff '+7D '+7h '+H '); Final predicted value d=Clip1 ((d '+64)＞＞7).Wherein ff ' and h ' are relevant position 1/2nd sample medians, and D ' and H ' are that the relevant position integral sample is amplified 8 times value.The interpolation process of/4th sample n is identical with the interpolation process of d.
/ 4th sample i: at first with F2 in the horizontal direction to gg ', h ", j ' and m " four value filterings, obtain median i '=(gg '+7h "+7j '+m "); Final predicted value i=Clip1 ((i '+512)＞＞10).Wherein gg ' and j ' they are relevant position 1/2nd medians, h " and m " be that relevant position 1/2nd sample medians are amplified 8 times value, the interpolation process of 1/4th sample k is identical with the interpolation process of i.
/ 4th sample f: at first with F2 in vertical direction to hh ', b ", j ' and s " four value filterings, obtain median f '=(hh '+7b "+7j '+s "); Final predicted value f=Clip1 ((f '+512)＞＞10).Wherein hh ' and j ' they are relevant position 1/2nd medians, b " and s " be that relevant position 1/2nd sample medians are amplified 8 times value, the interpolation process of 1/4th sample q is identical with the interpolation process of f.
/ 4th sample e, g, p and r:
e＝(D”+j’+64)＞＞7
g＝(E”+j’+64)＞＞7
p＝(H”+j’+64)＞＞7
r＝(I”+j’+64)＞＞7
D wherein ", E ", H " and I " be that the relevant position integral sample is amplified 64 times value, j ' is relevant position 1/2nd sample medians.
Consult Fig. 2, among this figure the whole location of pixels in the upper left corner of current luminance block E be (x, y), forecast sample matrix predMatrix[x, y] according to table 1 assignment.
Table 1 forecast sample matrix pixel table
| ????xFracL | ??0 | ??0 | ??0 | ??0 | ??1 | ??1 | ??1 | ??1 | ??2 | ??2 | ??2 | ??2 | ??3 | ??3 | ??3 | ??3 |
| ????yFracL | ??0 | ??1 | ??2 | ??3 | ??0 | ??1 | ??2 | ??3 | ??0 | ??1 | ??2 | ??3 | ??0 | ??1 | ??2 | ??3 |
| ????predMatrix ????[x，y] | ??D | ??d | ??h | ??n | ??a | ??e | ??I | ??p | ??b | ??f | ??j | ??q | ??c | ??g | ??k | ??R |
Annotate: xFracL equals the horizontal component mvE_x ﹠amp of mvE; 3, yFracL equals the vertical component mvE_y ﹠amp of mvE; 3
Degree of filling in colors on a sketch sample interpolation uses and the motion vector of corresponding brightness piece is mvE, and the horizontal component of mvE is mvE_x, and vertical component is mvE_y, and corresponding motion vector is mvC.The horizontal component of mvC is mvC_x, and vertical component is mvC_y, and the base unit of mvC is 1/8 sample.
Shown in Fig. 3 chroma sample interpolation graphs, A, B, C, D are by the integral sample value around the interpolated sample, and dx and dy are respectively level and the vertical ranges of forecast sample and A, and dx equals mvC_x ﹠amp; 7, dy equals mvC_y ﹠amp; 7, the plain predMatrix[x of forecast sample entry of a matrix, y] calculate according to following formula:
predMatrix[x，y]＝((8-dx)×(8-dy)×A+dx×(8-dy)×B+(8dx)×dy×C+dx×dy×D+32)＞＞6
See also Fig. 4, video decoding chip fraction pixel filtering system comprises memory, with the data transmission circuit of a register array, with the low pass filter of four data selectors.Memory belongs to peripheral interface, is provided by the external world.13 integer position pixels of delegation through the filter circuit filtering of band selector, are exported the fraction pixel position through data transmission circuit output 6 * 6 integer pixel positions matrixes of band register array.The data transmission circuit in the native system and the concrete function of low pass filter are as follows:
1, data transmission circuit
Data transmission circuit looks like to be connected with external memory, and the pixel of the whole picture element matrix that will receive from the external memory control unit interface deposits memory in the form of row.From top algorithm as can be seen, insert out whole location of pixels and be (i, j), the point of all fraction pixel positions, need one 6 * 6 whole picture element matrix, the position, the upper left corner of this matrix be (i-2, j-2), position, the lower right corner is (i+3, j+3), so the brightness that will insert out one 8 * 8 divides picture element matrix, need one 13 * 13 the whole picture element matrix of brightness, insert out one 4 * 4 colourity and divide picture element matrix, need one 5 * 5 the whole picture element matrix of colourity.
As shown in Figure 5, in chip, the reference data of whole pixel will be sent by the module of upstream in the memory RAM that is put on the sheet and deposit by row, and there are 13 pixels in delegation, and this can improve the utilization rate of hardware.
This is one 6 * 13 a register array, and register array one end connects data selector, and the whole picture element matrix of being sent by memory is moved in circulation, and the other end connects low pass filter.After coming out, data will enter this array from RAM, the unit of this array is divided into two classes, 6 * 6 top unit are active units, wherein depositing current 6 * 6 required matrix pixel of interpolation that carries out, in order to make 6 * 6 top matrixes is current required data all the time, and the data in this array have three kinds of move modes, move in the circulation, circulation moves down, and moves horizontally.
In i.e. 6 bats of 6 initial clock cycle after the interpolation of one 8 * 8 of startups, 13 pixels of delegation are taken out in every bat from RAM, obtain 6 * 13 initial each pixel of row in the mode that moves horizontally, and at this moment Shang Mian 6 * 6 elements are effective.Array will be done in the circulation and move in ensuing 7 clap, and it is effective to remain 6 * 6 top elements like this.Clap to do once again the 8th and move horizontally, new one-row pixels is come to shift-in, but because the array of this moment moves owing to having done in 7 circulations, new one-row pixels puts in order when coming in must do once and is inverted, 7 pixels of back are moved to the front, and 6 pixels of front are moved to the back.6 * 6 elements above in the array are again effectively after the 8th claps like this, and last image element interpolation that only current active data is second row is needed, and array will be done circulation and move down then, move horizontally after clapping until 7 again.And then do in the circulation and move, every like this bat can both sent 6 * 6 whole pixels.
For chrominance block, can multiplexing this structure, array is also done above-mentioned three kinds of actions, and only the required data of interpolation will be positioned at 2 * 2 elements in the upper left corner.
2, based on the filter circuit of streamline
Low pass filter receives the whole picture element matrix pixel of input, and the integer pixel matrix pixel is carried out 1/2 and 1/4 filtering, obtains the fraction pixel position, and its filtering output end connects the whole picture element matrix in the memory, and the fraction pixel position is inserted in the whole picture element matrix.Filter mainly contains two kinds:
(1) luminance filter.As shown in Figure 6, in order to reduce the length of critical path, one 1/2 luminance filtering is splitted into two clap and finish, multiplication becomes displacement and addition, as a * 5=a＜＜2+a, the structure of 1/4 filtering is similar with it, but the coefficient difference, h represents 1/2 filtering, q represents 1/4 filtering.
Clap outflow one number for each, the present invention adopts the pipeline organization design, for the different pipeline organization of some employing of diverse location.
For the 1st class point, need 1 1/2 filter, drawing a number needs 2 to clap altogether, and for example b only need can draw by 1/2nd picture element interpolations of a sub-level, and i is similar with it, is finished by 1/2 filter h0.Structure as shown in Figure 6.
For the 2nd class point, need 2 1/2 filters and one 1/4 filter, drawing a number needs 4 to clap altogether, two level 1/4 pixels of a, c need to draw two 1/2 pixels by 1/2 interpolation of two sub-levels, this is by 1/2 filter h0, and h1 finishes, and 1/4th filtering that utilize these two 1/2 pixels and two whole pixels to carry out a sub-level then draw, this is finished by 1/4 filter q0, and d, n are similar with it.Structure as shown in Figure 7.
For the 3rd class point, need 5 1/2 filters and (or) a weighted average device, need 4 to clap or 5 bats altogether, 1/2 filtering that j need carry out four sub-levels earlier obtains four 1/2 pixels, this filtering is finished by 1/2 filter h0, h1, h2, h3, carry out once 1/2 vertical filtering with these four pixels again and just can obtain, promptly finished by 1/2 filter h4, e, g, four 1/4 need of p, r are weighted j on average and can get with the whole pixel of corresponding diagonal.Structure as shown in Figure 8.
For the 4th class point, for these two 1/4 pixels of f, q, need carry out 1/2 interpolation of five sub-levels earlier, finish by 1/2 filter h0, h1, h2, h3, h4, utilize these five/2nd pixels to carry out 1/2 vertical interpolation twice then, this step is finished by 1/2 filter h5, h6.Utilize the result of this twice interpolation and two whole pixels to carry out one time 1/4th interpolation again and obtain, finished by 1/4 filter q0, the producing method of i, k is similar with it, only the horizontal filtering and the direction of vertically filtering has been exchanged.Need 7 1/2 filters and one 1/4 filter, need 6 to clap altogether, structure as shown in Figure 9.
In fact, for the required filter of first three class point streamline, all can provide, therefore can finish the computing work of all types point, just need carry out different configurations to streamline according to different positions with the 4th kind of streamline by the streamline of the 4th class point.Whole logics of configuration all embody in 4 selector Muxer, and its structure as shown in figure 10.
The working method of selector is that its working method is described below according to the different flow direction of the different determination datas of dx and dy: the right-hand member of arrow is the input of selector in the following description, and left end then is output.
Selector 0:
h0_i0←A11
h0_i1←A12
h0_i2←A13
h0_i3←A14
h1_i0←A21
h1_i1←A22
h1_i2←A23
h1_i3←A24
h2_i0←A31
h2_i1←A32
h2_i2←A33
h2_i3←A34
h3_i0←A41
h3_i1←A42
h3_i2←A43
h3_i3←A44
If dx equal 1 and dy equal 1, so
A0_i←A22＜＜6
Else if dx equal 1 and dy equal 3, so
A0_i←A23＜＜6
Else if dx equal 3 and dy equal 1, so
A0_i←A323＜＜6
Else if dx equal 3 and dy equal 3, so
A0_i←A333＜＜6
If dx equal 1 and dy equal 0, so
h4_i0←A20；
h4_i1←A21；
h4_i2←A22；
h4_i3←A23；
Else if dx equal 3 and dy equal 0, so
h4_i0←A22
h4_i1←A23
h4_i2←A24
h4_i3←A25
Else if dx equal 2 and dy equal 1, so
h4_i0←A11
h4_i1←A12
h4_i2←A13
h4_i3←A14
Else if dx equal 2 and dy equal 3, so
h4_i0←A51
h4_i1←A52
h4_i2←A53
h4_i3←A54
Selector 1:
h5_i0←h0_O
h5_i1←h1_O
h5_i2←h2_O
h5_i3←h3_O
If dy equals 1, so
h5_i0←h4_O
h5_i1←h0_O
h5_i2←h1_O
h5_i3←h2_O
Otherwise
h5_i0←h1_O
h5_i1←h2_O
h5_i2←h3_O
h5_i3←h4_O
If dx equals 1, so
q0_i0←h4_O
q0_i1←A22′＜＜3
q0_i2←h1_O
q0_i3←h23′＜＜3
Otherwise
q0_0←A22′＜＜3
q0_i1←h1_O
q0_i2←A23′＜＜3
q0_i3←h4_O
Selector 2:
If dy equals 1, so
q1_i1←h6；
q1_i2←h1′＜＜3；
q1_i3←h5；
q1_i4←h2′＜＜3；
Otherwise
q1_i1←h1′＜＜3；
q1_i2←h5；
q1_i3←h2′＜＜3；
q1_i4＝h6；
Selector 3:
If this point is the integer pixel point
result←clip1((h1_O+4)＞＞3)；
If this point is the 1st a class point
result←clip1((q0_O+64)＞＞7)；
If this point is the 2nd a class point
reslut←clip1((h5_O+32)＞＞6)；
If this point is the 3rd a class point
result←clip1((A_O+64)＞＞7)；
If this point is the 4th a class point
result←clip1((q1_O+512)＞＞10)；
(2) chrominance filter.As shown in figure 11.The structure of chrominance filter is single, in order to shorten critical path, a chroma interpolation is divided into 4 bats finishes.
Above embodiment is the unrestricted technical scheme of the present invention in order to explanation only, those of ordinary skill in the art is to be understood that: can make amendment or be equal to replacement the present invention, and not breaking away from any modification or partial replacement of the spirit and scope of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (17)
1, video decoding chip fraction pixel filtering system, it is characterized in that, this system comprises memory, the data transmission circuit of band register array, the low pass filter of band data selector, low pass filter is based on the low pass filter of streamline, data transmission circuit is connected with external memory, to deposit register array in the form of going from the whole picture element matrix that the external memory control unit interface receives, the whole location of pixels of being sent by memory is moved in the register array circulation, and the whole pixel of the activity after will circulating is sent into low pass filter, low pass filter carries out 1/2 and 1/4 filtering, data selector is exported the fraction pixel position according to the type of flow of the Position Control data of fraction pixel.
2, video decoding chip fraction pixel filtering system according to claim 1 is characterized in that, described data selector is 4.
3, video decoding chip fraction pixel filtering system according to claim 1 is characterized in that, described whole picture element matrix is the whole picture element matrix of one 13 * 13 brightness.
4, video decoding chip fraction pixel filtering system according to claim 1 is characterized in that, described whole picture element matrix is 5 * 5 the whole picture element matrix of colourity.
5, video decoding chip fraction pixel filtering system according to claim 1 is characterized in that described register array is divided into two parts, and wherein top unit is active unit all the time, is 6 * 13 matrix.
6, video decoding chip fraction pixel filtering system according to claim 1 is characterized in that, the whole pixel of described activity is one 6 * 6 a matrix.
7, video decoding chip fraction pixel filtering system according to claim 1 is characterized in that, described low pass filter is 4 tap F
1(1,5,5 ,-1) filter is predicted 1/2 sample position of nearest pixel samples.
8, video decoding chip fraction pixel filtering system according to claim 1 is characterized in that, described low pass filter is 4 tap F
2(1,7,7,1) filter is predicted 1/4 sample position of nearest pixel samples.
9, video decoding chip fraction pixel filtering method is characterized in that, this method may further comprise the steps at least:
Step 1, data transmission circuit then will deposit in the memory with the form of going from the whole pixel that the external memory control interface receives;
Step 2, data selector move into register array with the whole pixel in the memory with matrix form;
The pixel of the whole picture element matrix of being sent by memory is moved in step 3, register array circulation, and will send into low pass filter through the pixel of the mobile movable integer pixel matrix that obtains of circulation;
Step 4, low pass filter carry out 1/2 and 1/4 filtering to the pixel of movable integer pixel matrix, obtain the fraction pixel position;
Step 5, the fraction pixel position is exported.
10, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that, the register array circulation in the described step 3 is moved and be may further comprise the steps:
Step 31, in beginning 6 clock cycle, each cycle is taken out delegation's totally 13 pixels from memory, move into register array in the mode of moving horizontally;
Step 32, in 7 cycles next, array will be done in the circulation and move, it is effective to remain 6 * 6 top pixels;
Step 33, move into new one-row pixels with the inverted level of putting in order, 7 pixels of back are moved to the front in the 8th cycle;
Step 34, array are done circulation and are moved down;
Step 35,7 cycles move horizontally later again;
Whether the whole pixel of step 36, judgement input has moved, and intact then end is moved, otherwise repeating step 1.
11, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that the filtering in the described step 4 specifically may further comprise the steps:
Step 411, use F
1, F
24 integral sample filtering to nearest on level or the vertical direction obtain median;
Step 412, median are through after moving to right, and if resulting value is greater than 255, and then final predicted value is 255, and less than 0, then final predicted value is 0 as if the value that obtains, otherwise final predicted value is the resulting value that moves to right.
12, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that described filtering specifically may further comprise the steps:
Step 421, use F
1, F
2Nearest 4 integers or fractional samples on the tilted direction are carried out 1/2 or 1/4 sample filtering, obtain median;
Step 422, median are through after moving to right, and if resulting value is greater than 255, and then final predicted value is 255, and less than 0, then final predicted value is 0 as if the value that obtains, otherwise final predicted value is the resulting value that moves to right.
13, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that, the filtering in the described step 4 adopts multiplication to become the method for displacement and addition, divides 2-7 clock cycle to finish.
14, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that, the filtering in the described step 4 is that the clock cycle is 2 with 1 1/2 filter filtering the 1st class point.
15, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that, the filtering in the described step 4 is that the clock cycle is 4 with 2 1/2 filters and 1 1/4 filter filtering the 2nd class point.
16, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that, the filtering in the described step 4 is that the clock cycle is 4 or 5 with 5 1/2 filters and 1 weighted average device filtering the 3rd class point.
17, video decoding chip fraction pixel filtering method according to claim 9 is characterized in that, the filtering in the described step 4 is that the clock cycle is 6 with 7 1/2 filters and 1 1/4 filter filtering the 4th class point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100703639A CN1306820C (en) | 2004-07-29 | 2004-07-29 | Fractional pixel filter system and its filter method for video frequency decoding chip |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100703639A CN1306820C (en) | 2004-07-29 | 2004-07-29 | Fractional pixel filter system and its filter method for video frequency decoding chip |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1589020A true CN1589020A (en) | 2005-03-02 |
| CN1306820C CN1306820C (en) | 2007-03-21 |
Family
ID=34604448
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100703639A Active CN1306820C (en) | 2004-07-29 | 2004-07-29 | Fractional pixel filter system and its filter method for video frequency decoding chip |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1306820C (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100394797C (en) * | 2006-04-13 | 2008-06-11 | 上海交通大学 | Method of realizing VLSI of brightness interpolator based on AVS movement compensation |
| CN100455025C (en) * | 2005-07-21 | 2009-01-21 | 华为技术有限公司 | Fractional pixel interpolation method and encoding and decoding apparatus |
| CN100576918C (en) * | 2007-06-27 | 2009-12-30 | 中国科学院微电子研究所 | A kind of AVS interframe predicting pixel generating apparatus |
| CN104811739A (en) * | 2015-04-27 | 2015-07-29 | 重庆邮电大学 | H.264 rapid interpolation method based on movement compensation |
| CN106507118A (en) * | 2016-11-28 | 2017-03-15 | 济南浪潮高新科技投资发展有限公司 | A kind of bimodulus brightness interpolating filter structure and method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR0159559B1 (en) * | 1994-10-31 | 1999-01-15 | 배순훈 | Adaptive postprocessing method of a digital image data |
| US6259734B1 (en) * | 1998-04-03 | 2001-07-10 | Matsushita Electric Industrial Co., Ltd. | Image processing method, image processing apparatus, and data storage media |
| US6859554B2 (en) * | 2001-04-04 | 2005-02-22 | Mitsubishi Electric Research Laboratories, Inc. | Method for segmenting multi-resolution video objects |
| ATE454016T1 (en) * | 2001-09-12 | 2010-01-15 | Panasonic Corp | IMAGE CODING AND DECODING METHODS |
| US7630566B2 (en) * | 2001-09-25 | 2009-12-08 | Broadcom Corporation | Method and apparatus for improved estimation and compensation in digital video compression and decompression |
-
2004
- 2004-07-29 CN CNB2004100703639A patent/CN1306820C/en active Active
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100455025C (en) * | 2005-07-21 | 2009-01-21 | 华为技术有限公司 | Fractional pixel interpolation method and encoding and decoding apparatus |
| CN100394797C (en) * | 2006-04-13 | 2008-06-11 | 上海交通大学 | Method of realizing VLSI of brightness interpolator based on AVS movement compensation |
| CN100576918C (en) * | 2007-06-27 | 2009-12-30 | 中国科学院微电子研究所 | A kind of AVS interframe predicting pixel generating apparatus |
| CN104811739A (en) * | 2015-04-27 | 2015-07-29 | 重庆邮电大学 | H.264 rapid interpolation method based on movement compensation |
| CN106507118A (en) * | 2016-11-28 | 2017-03-15 | 济南浪潮高新科技投资发展有限公司 | A kind of bimodulus brightness interpolating filter structure and method |
| CN106507118B (en) * | 2016-11-28 | 2019-10-11 | 浪潮集团有限公司 | A kind of bimodulus brightness interpolating filter structure and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1306820C (en) | 2007-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1199471C (en) | Image, sound processing device of improving process efficiency | |
| CN1238798C (en) | Double cubic filter | |
| CN1225126C (en) | Space predicting method and apparatus for video encoding | |
| CN1662068A (en) | Moving image encoding apparatus and moving image processing apparatus | |
| CN1816149A (en) | Filtering-wave method of removing blocking-effect in video-frequency picture and loop filter | |
| CN1191513C (en) | Display driver, display device and electronic equipment having said display driver | |
| CN1256849C (en) | Method and apparatus for 1/4 pixel precision interpolation | |
| CN1456999A (en) | Color movement vector integerization | |
| CN1774071A (en) | Apparatus for motion estimation using a two-dimensional processing element array and method therefor | |
| CN1131871A (en) | Moving picture processing method | |
| CN1119386A (en) | Picture coding and decoding apparatus | |
| CN1469632A (en) | Video frequency coding/decoding method and equipment | |
| CN1449198A (en) | Moving picture coding method, moving picture coding apparatus, and moving picture coding program | |
| CN1428939A (en) | Video coder | |
| CN1453725A (en) | Round-off control for multi-stage interpolation | |
| CN1909666A (en) | Motion estimation apparatus | |
| CN1816154A (en) | Method and apparatus for motion estimation | |
| CN1685369A (en) | Low complexity and unified transforms for video coding | |
| CN1845213A (en) | Method for realizing encryption/decryption processing in SMS4 cipher algorithm | |
| CN1270541C (en) | Decoder and method thereof, coding device and method thereof, image processing system and method thereof | |
| CN1652608A (en) | Data processing device and method of same, and encoding device and decoding device | |
| CN1240225C (en) | Picture coding device and picture coding method | |
| CN1321273A (en) | Operating system and virtual computer system | |
| CN1306820C (en) | Fractional pixel filter system and its filter method for video frequency decoding chip | |
| CN1529509A (en) | Video image sub-picture-element interpolation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: ZHANXUN COMMUNICATIONS (SHANGHAI) CO., LTD. Free format text: FORMER OWNER: UNITED XINYUAN DIGITAL AUDIO-VIDEO TECHNOLOGY (BEIJING) CO., LTD. Effective date: 20070608 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20070608 Address after: 201203 Shanghai city Zuchongzhi road Pudong Zhangjiang hi tech park, Spreadtrum Center Building 1, Lane 2288 Patentee after: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Address before: 100080 North building, room 6, 140 South Road, Haidian District Academy of Sciences, Beijing Patentee before: UNITED XINYUAN DIGITAL AUDIO V |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190313 Address after: 101399 Building 8-07, Ronghui Garden 6, Shunyi Airport Economic Core Area, Beijing Patentee after: Xin Xin finance leasing (Beijing) Co.,Ltd. Address before: 201203 Shanghai city Zuchongzhi road Pudong Zhangjiang hi tech park, Spreadtrum Center Building 1, Lane 2288 Patentee before: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20050302 Assignee: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Assignor: Xin Xin finance leasing (Beijing) Co.,Ltd. Contract record no.: X2021110000008 Denomination of invention: Fractional pixel filtering system and filtering method for video decoder chip Granted publication date: 20070321 License type: Exclusive License Record date: 20210317 |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221025 Address after: 201203 Shanghai city Zuchongzhi road Pudong New Area Zhangjiang hi tech park, Spreadtrum Center Building 1, Lane 2288 Patentee after: SPREADTRUM COMMUNICATIONS (SHANGHAI) Co.,Ltd. Address before: 101399 Building 8-07, Ronghui Garden 6, Shunyi Airport Economic Core Area, Beijing Patentee before: Xin Xin finance leasing (Beijing) Co.,Ltd. |