CN100455002C - Method and circuit for zooming digital video image based on dual quadratic interpolation - Google Patents
Method and circuit for zooming digital video image based on dual quadratic interpolation Download PDFInfo
- Publication number
- CN100455002C CN100455002C CNB2005100962924A CN200510096292A CN100455002C CN 100455002 C CN100455002 C CN 100455002C CN B2005100962924 A CNB2005100962924 A CN B2005100962924A CN 200510096292 A CN200510096292 A CN 200510096292A CN 100455002 C CN100455002 C CN 100455002C
- Authority
- CN
- China
- Prior art keywords
- data
- field signal
- input
- interpolation
- read
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
The present invention discloses a method and a circuit for scaling a digital video image based on dual quadratic interpolation. YUV format or RGB format digital video data with 24 bits are written into a memory under the function of an input line field signal, and a module is then generated by the input line-field signal to regulate and generate an output line-field signal. A reading address of a data read-out memory is generated by the accumulated mapping calculation of a scaling factor under the function of the output line-field signal, and data after interpolation is obtained from read-out data by weighting calculation. The realizing circuit of the method is simple, and amplification and reduction of the image can be realized in a same circuit. Bicubic interpolation algorithm of image scaling of a digital video is realized by the generating method of the output line-field signal and the accumulated mapping calculation of the scaling factor through the read-write control logics of few hardware resources and the simple data memory. Meanwhile, the method and the circuit of the present invention can also be used as an independent IP core, and apply to various occasions which need the scaling of the video images.
Description
Technical field
The invention belongs to digital video image and handle and the Display Technique field, be specifically related to a kind of method and circuit thereof of the zooming digital video image based on bicubic interpolation.
Background technology
Zooming digital video image chip (Scaler) is the indispensable part of flat-panel display device.Because flat-panel display device has intrinsic resolution, promptly the resolution of a flat-panel display device is fixed, thus the different video source of resolution on flat-panel display device, show, must be the intrinsic resolution of flat-panel display device with its conversion of resolution.The zooming digital video image chip promptly is to realize such image resolution ratio translation function, and the image zoom interpolation algorithm is the core algorithm of zooming digital video image chip.
The image zoom interpolation algorithm is commonly used arest neighbors interpolation, bilinear interpolation and bicubic interpolation, wherein the effect of bicubic interpolation is best, the high-frequency information that can keep image preferably, the image border is sharper keen, details is more clear, improved the mosaic phenomenon of arest neighbors interpolation and the problem of image blurring of bilinear interpolation, but amount of calculation and hardware implementation complexity are more much bigger than preceding two kinds of algorithms.
Summary of the invention
Defective or deficiency at above-mentioned prior art existence, the objective of the invention is to, a kind of method and device thereof of the zooming digital video image based on bicubic interpolation are proposed, this method can be saved hardware resource, need not the fifo buffer memory before and after the data storage, just can guarantee can not take place read-write and catch up with problem, the memory read write control logic is simple.The realization circuit of this method is simple, in same circuit, can realize the amplification of image and dwindles.
In order to achieve the above object, technical scheme of the present invention is, the digital of digital video data of 24 yuv formats or rgb format is write entry data memory under the effect of line of input field signal, produce output row/field signal by the modulation of line of input field signal again, the data read data memory read the add up mapping operations generation of address by the scaling factor under output row/field signal effect, the data based bicubic interpolation computing formula of reading computes weighted and draws data after the interpolation.This process both had been applicable to that vertically scale also was applicable to horizontal scaling.Wherein most critical is the generation of output row field signal and the mapping operations that adds up of the scaling factor.
A kind of circuit of realizing said method is characterized in that, this circuit comprises:
One data storage, the bit wide of data storage is 24, is used to store 24 the yuv format or the digital of digital video data of rgb format, data serial input, and line output, each 4 data of parallel read-out;
Delegation/field signal generation module is used to guarantee that data write, the synchronism of read data memory;
One writes control module, be used under the effect of line of input field signal, by to line of input/counting, that gets count results hangs down 3, produce writing of data storage and enable and write address, the digital of digital video data of yuv format or rgb format is write entry data memory according to row/piont mark serial successively;
One convergent-divergent mapping block is used to finish the mapping that adds up of the scaling factor and calculates; Obtain representing the range information d value between the picture element in the positional information i value of picture element in the required original image of current interpolation point and current interpolation point and the required original image;
One reads control module, the positional information i value that the original image picture element that provides by the convergent-divergent mapping block is provided produce data storage read to enable and read current interpolation point is calculated 4 required data from data storage, reads simultaneously in the address;
One interpolation calculation module is weighted calculating according to the bicubic interpolation computing formula, receives by reading data that control module sends into and the range information d value of being sent here by the convergent-divergent mapping block, calculating interpolation result;
Under line of input/field signal effect, produce to write and enable and write address by writing control module, the digital of digital video data of 24 yuv formats or rgb format is write entry data memory; OK/the field signal generation module produces the output row field signal by line of input field signal modulation; Under the effect of output row/field signal, the convergent-divergent mapping block is finished the mapping operations that adds up to the scaling factor, produces the range information d value between the picture element in the positional information i value of picture element in the required original image of current interpolation point and current interpolation point and the required original image; Reading positional information i value that control module sent here by the convergent-divergent mapping block produces and reads to enable and read the address, from data storage, read 4 required data simultaneously, send into the interpolation calculation module, the range information d value that the interpolation calculation module is sent here in conjunction with the convergent-divergent mapping block, ranking operation draws the data after the interpolation.
The present invention realizes the bicubic interpolation algorithm of zooming digital video image by designing the mapping operations that adds up dexterously with exporting the row field signal production method and the scaling factor with the read-write control logic of less hardware resource, simple data storage.Simultaneously, the zooming digital video image device that method of the present invention realizes also can be used as independently IP kernel use, is applied to the various occasions that need video image zooming, as video format scanning chip.Its convergent-divergent multiple can reach between (1/4～4), and various VGA signals and high definition TV signal are supported in input, supports that input resolution can be up to 1280 * 1024, and output image resolution can reach 1280 * 1024, and clock frequency can reach 108M.
Description of drawings
Fig. 1 is the hardware implementation structure figure of one dimension directional interpolation of the present invention;
Data memory structure schematic diagram when Fig. 2 is vertically scale;
Data memory structure schematic diagram when Fig. 3 is horizontal scaling;
Fig. 4 is the location diagram of each point in vertical direction in the computing formula of bicubic interpolation of one dimension.
Below in conjunction with accompanying drawing the present invention is further described in detail.
Embodiment
Its principle of method of zooming digital video image based on bicubic interpolation of the present invention is as follows:
1, the generation of output row field signal
If the scaling factor is m/n, realize for the ease of hardware, require this scale factor to be the simplest prime number ratio, and m, n＜64.Satisfy and need not the fifo buffer memory before and after the data storage and can guarantee not take place to read and write the problem of catching up with, simplify the memory read write control logic, require output row field signal characteristic to be: (1) periodic row field signal, (2) field frequency is constant, input, differ the time delays of input 3 row between the output field useful signal, and line of input reach synchronously the capable time of useful signal m with the output row synchronously and the capable time of useful signal n equate.
If the every capable transmission time of input, output, synchronous signal cycle is respectively hperiod_in and hperiod_out at once, import the total pixel of every row and count and be htotal_in, and input clock cycle Tclk, then their pass is
Therefore as long as to input clock counting, every input
The time of individual picture element is exactly the time of output delegation, can generate the synchronous and useful signal of output row by the counting to the output clock; Output field reaches useful signal synchronously by the counting of exporting capable synchronous head is produced, and its cycle is identical with the input field sync signal cycle, differs the delay of input 3 row on the time.
For eliminating add up error the output row field signal is periodically influenced, the every input of above-mentioned line of input counting operation m is capable to be that zero clearing restarts new round count cycle, the last column that like this accumulated counts error is only existed each circulation, but this error is an acceptable for flat-panel display device.
2, the mapping operations that adds up of the scaling factor
Replace multiplication with accumulating operation, hardware realize to be simplified, and a length of the cycle of accumulating operation is controlled at 63 in (OK/point), has eliminated the influence of add up error to operation result substantially.
Bicubic interpolation is to utilize current interpolation point corresponding to 16 pixel point values in 4 neighborhoods around the original image position, is weighted calculating according to current interpolation point and their position relation, obtains new pixel point value.We are divided into vertical direction and twice one-dimensional operation of horizontal direction to the calculating of two dimension, can reduce computation complexity, simplify hardware and realize.
The computing formula of the bicubic interpolation of one dimension is as follows:
V[I，j]＝-d(1-d)
2V[i-1，j]+(1-2d
2+d
3)V[i，j]
+d(1+d-d
2)V[i+1，j]-d
2(1-d)V[i+2，j]
Wherein, V[*, *] be that coordinate is the pixel value (being YUV value or rgb value) of the point of [*, *], d is the distance of point [I, j] and point [i, j].If vertical direction, the position of each point concerns as shown in Figure 4 in the formula, as long as find four picture elements wherein and determine that the d value can calculate the pixel value of interpolation point, and this process promptly is to calculate by the mapping that adds up of the scaling factor to finish.
If former image is A, the image behind the convergent-divergent is B.Certain row pixel lattice point A among the intercepting A
1To Am, corresponding to B among the B
1To Bn.Interpolation calculation is a circulation with n the point (being m the point of A) of B, can eliminate the influence of add up error to result of calculation basically like this.
Make the area equal and opposite in direction of convergent-divergent front and back A and B, then A
1To Am and B
1Equal in length to Bn.If the pixel lattice point length of side of A is 1, then the pixel lattice point length of side of B is m/n, and is special, is positioned at the picture element A at two ends
1, Am and B
1, Bn is respectively 1/2 and m/2n apart from the edge.
In A, set up a coordinate system, A
1Coordinate to Am is 1 to m.B
IBe B
1To certain point, then B between the Bm
ICoordinate in A is
The integer part of the x as a result of following formula is exactly B
IA among the corresponding A
iThe coordinate figure of point, just i value; Fractional part is exactly B
IWith A
iIn coordinate system apart from the d value.
Realize for the ease of hardware,
Multiplying by
The accumulating operation of the factor replaces, and exports every row during vertically scale and adds up once, and every some during horizontal scaling (each exports clock) adds up once.The present invention is dexterously with this additive factor
Change into 12 the binary number that computer is easy to represent:
Wherein, high 6 bit representation integer parts, low 6 bit representation fractional parts.Like this, high 6 of the accumulation result of bit_rate promptly is the i value, and low 6 promptly is the d value divided by 64.
But because this approximate representation form of additive factor can cause error to accumulation calculating, accumulative frequency is many more, and error is big more.So be a circulation when the present invention arrives accumulative frequency in n, every n that promptly adds up is capable/zero clearing once, restart to add up.Add up error can not surpass integer 1 like this, can offset the influence of add up error to the result basically.
The circuit structure of one dimension direction of realizing said method is referring to Fig. 1, and it comprises: data storage 1, row field signal generation module 2, write control module 3, convergent-divergent mapping block 4, read control module 5, interpolation calculation module 6.Details are as follows:
1, data storage
The clock frequency of transfer of data will change before and after the convergent-divergent, so data must be passed through data storage.Data storage is that the row that 4 registers group constitute is deposited during the vertical direction convergent-divergent, referring to Fig. 2, and every row storage two row.Data storage is that the point that 8 triggers constitute is deposited during the horizontal direction convergent-divergent, referring to Fig. 3.Because the input data are 24 yuv format or rgb format, so the bit wide of memory is 24, and the capacity of registers group is according to the maximum horizontal size decision of digital video image signal.
Data serial input, and line output, each 4 data of parallel read-out.
2, row field signal generation module
Certain synchronous head trailing edge by the input field synchronization begins to trigger, and all operations all carries out when triggering signal is effective, guarantees the synchronism of input and output row field signal;
Line of input is counted synchronously, m capable synchronous head of every input (
Be the scaling factor) produce a pulse, generate the circulation reset signal, the pulse that following accumulated counts operation runs into this signal is that zero clearing restarts new round circulation;
To input clock counting, every input
Individual picture element produces a synchronous reset signal of output row, to exporting clock count zero clearing under this signal effect, begins to reach useful signal synchronously with end position generation output row by low level is set then;
Produce a pulse by certain the output row synchronous head trailing edge behind the input field synchronization head, generate the synchronous reset signal of output field, the output line synchronizing signal is gone counting, low level begins and end position by being provided with, and consider that the output field useful signal will lag behind time of input useful signal 3 row inputs, produces output field and reaches useful signal synchronously.
The output row field signal that meets the demands that produces is sent into the mapping operations that adds up that the convergent-divergent mapping block is used for the scaling factor.
3, write control module
Under the effect of line of input field signal, by to line of input/counting, get the low 3 of count results, produce writing of data storage and enable and write address, with data according to the row/piont mark among Fig. 2 and Fig. 3 serial write memory successively.
4, convergent-divergent mapping block
This module is finished the crucial scaling factor mapping computational process that adds up.Under the effect of output row field signal, variable bie_rate exports every row/add up once, every n that adds up is capable/zero clearing once, the positional informations (i value) of picture element in the required original image of high 6 current interpolation points of representative of the adder as a result that obtains are sent into and are read control module and be used for generation and read the address; Range information (d value) in low 6 current interpolation points of representative and the required original image between the picture element is sent into the interpolation calculation module and is used for weighted calculation.
5, read control module
By the i value that the convergent-divergent mapping block is sent here, get that it is low 3, what produce data storage reads to enable and read the address, desired data is read sent into the interpolation calculation module.
6, interpolation calculation module
According to the interpolation calculation formula,, calculate interpolation result by data and the d value that front-end module is sent here.Four calculating about the multinomial coefficient of d in the formula realize by look-up table.
Claims (4)
1. method based on the zooming digital video image of bicubic interpolation, it is characterized in that, the digital of digital video data of 24 yuv formats or rgb format is write entry data memory under the effect of line of input field signal, produce output row/field signal by the modulation of line of input field signal again, the data read data memory read the add up mapping operations generation of address by the scaling factor under output row/field signal effect, the data based bicubic interpolation computing formula of reading computes weighted and draws data after the interpolation.
2. a circuit of realizing the described method of claim 1 is characterized in that, this circuit comprises:
One data storage (1), the bit wide of data storage (1) is 24, is used to store 24 the yuv format or the digital of digital video data of rgb format, data serial input, and line output, each 4 data of parallel read-out;
Delegation/field signal generation module (2) is used to guarantee that data write, the synchronism of read data memory (1);
One writes control module (3), be used under the effect of line of input field signal, by to line of input/counting, that gets count results hangs down 3, produce writing of data storage (1) and enable and write address, the digital of digital video data of yuv format or rgb format is write entry data memory (1) according to row/piont mark serial successively;
One convergent-divergent mapping block (4) is used to finish the mapping that adds up of the scaling factor and calculates; Obtain representing the range information d value between the picture element in the positional information i value of picture element in the required original image of current interpolation point and current interpolation point and the required original image;
One reads control module (5), the positional information i value that the original image picture element that provides by convergent-divergent mapping block (4) is provided produce data storage (1) read to enable and read current interpolation point is calculated 4 required data from data storage (1), reads simultaneously in the address;
One interpolation calculation module (6) is weighted calculating according to the bicubic interpolation computing formula, receives by reading data that control module (5) sends into and the range information d value of being sent here by convergent-divergent mapping block (4), calculating interpolation result;
Under line of input/field signal effect, produce to write and enable and write address by writing control module (3), the digital of digital video data of 24 yuv formats or rgb format is write entry data memory (1); OK/field signal generation module (2) produces the output row field signal by line of input field signal modulation; Under the effect of output row/field signal, convergent-divergent mapping block (4) is finished the mapping operations that adds up to the scaling factor, produces the range information d value between the picture element in the positional information i value of picture element in the required original image of current interpolation point and current interpolation point and the required original image; Reading positional information i value that control module (5) sent here by convergent-divergent mapping block (4) produces and reads to enable and read the address, from data storage (1), read 4 required data simultaneously, send into interpolation calculation module (6), the range information d value that interpolation calculation module (6) is sent here in conjunction with convergent-divergent mapping block (4), ranking operation draws the data after the interpolation.
3. circuit as claimed in claim 2 is characterized in that, described row/field signal generation module specific implementation is:
Certain synchronous head trailing edge by the input field synchronization begins to trigger, and all operations all carries out when triggering signal is effective, guarantees the synchronism of input and output row field signal;
Line of input is counted synchronously, and m capable synchronous head of every input produces a pulse, generates the circulation reset signal,
Be the scaling factor, the pulse that following accumulated counts operation runs into this signal is that zero clearing restarts new round circulation;
To input clock counting, every input
Individual picture element produces a synchronous reset signal of output row, wherein htotal_in is the total pixel number of the every row of input, to the zero clearing under this signal effect of output clock count,, low level begins to produce the synchronous and useful signal of output row then with end position by being set;
Produce a pulse by certain the output row synchronous head trailing edge behind the input field synchronization head, generate the synchronous reset signal of output field, the output line synchronizing signal is gone counting, low level begins and end position by being provided with, and consider that the output field useful signal will lag behind time of input useful signal 3 row inputs, produces output field and reaches useful signal synchronously;
The output row field signal that meets the demands that produces is sent into the mapping operations that adds up that convergent-divergent mapping block (4) is used for the scaling factor.
4. circuit as claimed in claim 2 is characterized in that, described convergent-divergent mapping block (4) specific implementation is:
Finish the scaling factor mapping computational process that adds up, under output row field signal effect, variable bit_rate exports every row/add up once, and wherein bit_rate represents the 12 bit representation m/n*64 that transformed by scaling factor m/n, every n that adds up is capable/and a zero clearing is once
Be the scaling factor, the positional information i values of picture element in high 6 required original images of the current interpolation points of representative of the adder as a result that obtains are sent into and are read control module and be used for generation and read the address; Range information d value in low 6 current interpolation points of representative and the required original image between the picture element is sent into the interpolation calculation module and is used for weighted calculation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100962924A CN100455002C (en) | 2005-11-03 | 2005-11-03 | Method and circuit for zooming digital video image based on dual quadratic interpolation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100962924A CN100455002C (en) | 2005-11-03 | 2005-11-03 | Method and circuit for zooming digital video image based on dual quadratic interpolation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1761312A CN1761312A (en) | 2006-04-19 |
| CN100455002C true CN100455002C (en) | 2009-01-21 |
Family
ID=36707229
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100962924A Expired - Fee Related CN100455002C (en) | 2005-11-03 | 2005-11-03 | Method and circuit for zooming digital video image based on dual quadratic interpolation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100455002C (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101662598B (en) * | 2008-08-26 | 2013-04-24 | 深圳艾科创新微电子有限公司 | Scaling system for continuous video data stream |
| CN101616249B (en) * | 2009-07-28 | 2011-08-10 | 谭洪舟 | Video scaling device based on bicubic interpolation |
| CN102457722B (en) * | 2010-10-26 | 2014-01-22 | 珠海全志科技股份有限公司 | Processing method and device for Bayer image |
| CN103500435B (en) * | 2013-09-11 | 2016-06-08 | 西安交通大学 | A kind of adapting to image interpolation method of edge guiding and VLSI implement device thereof |
| CN105427245B (en) * | 2015-11-06 | 2018-12-07 | 浙江宇视科技有限公司 | A kind of bicubic interpolation error accumulation removing method and device |
| CN109873684B (en) * | 2017-12-01 | 2021-05-18 | 联发科技股份有限公司 | Signal receiving device and signal processing method thereof |
| CN111199515B (en) * | 2018-11-16 | 2024-03-19 | 西安诺瓦星云科技股份有限公司 | Scaler, image scaling method and device |
| CN109767380B (en) * | 2018-12-12 | 2023-01-13 | 中国航空工业集团公司西安航空计算技术研究所 | Image zooming circuit and method for graphic processor |
| CN111770342B (en) * | 2020-06-19 | 2023-09-05 | 艾索信息股份有限公司 | Video stepless scaling method |
| CN114125182A (en) * | 2021-11-25 | 2022-03-01 | 中航华东光电有限公司 | Method for realizing 4K high-definition display based on FPGA |
| CN115150566A (en) * | 2022-09-01 | 2022-10-04 | 杭州雄迈集成电路技术股份有限公司 | Multi-path multi-resolution video real-time output method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1208306A (en) * | 1997-08-07 | 1999-02-17 | 三星电子株式会社 | Image format converting apparatus and methods in video signal processing system |
| US6229571B1 (en) * | 1998-07-23 | 2001-05-08 | Nec Corporation | Scan converter with interpolating function |
| US20020140853A1 (en) * | 2001-02-20 | 2002-10-03 | Seungjoon Yang | Sampling rate conversion apparatus and method thereof |
| US20020191114A1 (en) * | 2001-06-14 | 2002-12-19 | Ikuo Someya | Video signal processing circuit and method for converting number of scan lines and image display device using the same |
-
2005
- 2005-11-03 CN CNB2005100962924A patent/CN100455002C/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1208306A (en) * | 1997-08-07 | 1999-02-17 | 三星电子株式会社 | Image format converting apparatus and methods in video signal processing system |
| US6229571B1 (en) * | 1998-07-23 | 2001-05-08 | Nec Corporation | Scan converter with interpolating function |
| US20020140853A1 (en) * | 2001-02-20 | 2002-10-03 | Seungjoon Yang | Sampling rate conversion apparatus and method thereof |
| US20020191114A1 (en) * | 2001-06-14 | 2002-12-19 | Ikuo Someya | Video signal processing circuit and method for converting number of scan lines and image display device using the same |
Non-Patent Citations (1)
| Title |
|---|
| 一种视频图像分辨率提升的VLSI设计与实现. 冯华，徐维朴，姚霁，葛晨阳，柴屹华.微电子学与计算机，第21卷第1期. 2004 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1761312A (en) | 2006-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100455002C (en) | Method and circuit for zooming digital video image based on dual quadratic interpolation | |
| CN100552712C (en) | Image processing apparatus, image processing method and aberration emendation method | |
| EP2533192B1 (en) | Image processing apparatus, image processing method, and distortion correcting method | |
| US8749667B2 (en) | System and method for maintaining maximum input rate while up-scaling an image vertically | |
| CN101616249B (en) | Video scaling device based on bicubic interpolation | |
| CN102186044A (en) | Edge correlation image stepless scaling algorithm and hardware realization device thereof | |
| JP2014086845A (en) | Image processing apparatus and imaging apparatus | |
| CN103916612A (en) | Random proportion zoom system and method | |
| US7349032B2 (en) | Circuit to resize and enlarge an image signal and resizing and enlargement method for an image signal | |
| US7110620B2 (en) | Apparatus for processing digital image and method therefor | |
| JP2007079708A (en) | Image processor and processing method | |
| US20090091585A1 (en) | Screen enlargement/reduction device | |
| US10332239B2 (en) | Apparatus and method for parallel polyphase image interpolation | |
| CN101662598B (en) | Scaling system for continuous video data stream | |
| CN111107295A (en) | Video scaling method based on FPGA and nonlinear interpolation | |
| US20120050820A1 (en) | Image processing apparatus, control method of the same, and program | |
| US20090087119A1 (en) | Method and apparatus for arbitrary ratio image reduction | |
| CN100508550C (en) | Image-zooming system capable of saving memory | |
| US6806914B1 (en) | Method and apparatus for changing the size of image | |
| JP2000187726A (en) | Data interpolation method and its device and storage medium | |
| CN102129667B (en) | Image scaling method | |
| CN101984668B (en) | Real-time image scaling engine suitable for various 4*4 interpolation filters | |
| JP4465844B2 (en) | Image processing apparatus and method, and recording medium | |
| CN114741352B (en) | FPGA-based bilinear interpolation resampling implementation method and device | |
| JP4465570B2 (en) | Image processing apparatus and method, and recording medium |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090121 Termination date: 20141103 |
|
| EXPY | Termination of patent right or utility model |