CN111028129B - TLM microstructure for GPU pixel rectangular scaling and turning algorithm - Google Patents
TLM microstructure for GPU pixel rectangular scaling and turning algorithm Download PDFInfo
- Publication number
- CN111028129B CN111028129B CN201911125671.XA CN201911125671A CN111028129B CN 111028129 B CN111028129 B CN 111028129B CN 201911125671 A CN201911125671 A CN 201911125671A CN 111028129 B CN111028129 B CN 111028129B
- Authority
- CN
- China
- Prior art keywords
- row
- scaling
- module
- pixel
- coordinates
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Editing Of Facsimile Originals (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of computer hardware modeling, in particular to a TLM microstructure design for a GPU pixel rectangular scaling and turning algorithm. The TLM microstructure of the GPU-pixel-oriented rectangular scaling and flipping algorithm includes a calculate new map location module 1, an assign row pixel module 2, a process module 3 in the row direction, and a process module 4 in the column direction. The invention realizes the function and the realization structure of the pixel rectangle scaling and overturning algorithm based on the TLM model, realizes the function defined by the glPixelZoom () function defined by the OpenGL API, solves the problem of insufficient performance of scaling and overturning the pixel rectangle by the GPU hardware, and effectively accelerates the RTL design and development.
Description
Technical Field
The invention relates to the technical field of computer hardware modeling, in particular to a TLM microstructure for a GPU pixel rectangular scaling and turning algorithm.
Background
In the design and development of graphics processor chips (GPUs), the accuracy and efficiency of the algorithm are important factors in determining the functionality and performance of the GPUs. The glPixelZoom () function defined by the OpenGL API supports arbitrary enlargement, reduction, and flipping of an image, but does not define an algorithm for image scaling flipping. If the calculation amount of the implementation algorithm is large, and the structural division of the algorithm is unreasonable, the performance of the GPU for realizing scaling and overturning can be seriously reduced. Therefore, it is necessary to verify the algorithm and the algorithm-based structure as early as possible before the GPU chip hardware logic is implemented, and provide a reference basis for RTL design.
Disclosure of Invention
Based on the problems in the background technology, the TLM microstructure for the GPU pixel rectangle scaling and turning algorithm provided by the invention can solve the problems of accuracy and high efficiency of rtl simulation pixel rectangle scaling and turning algorithm and can perform function verification on a TLM model on the hardware microstructure of the pixel rectangle scaling and turning algorithm in advance of rtl.
The technical scheme of the invention is as follows:
a TLM microstructure for GPU-pixel-oriented rectangular scaling flip algorithm is provided, the structure comprising a calculate new map location module 1, an assign row pixel module 2, a process module 3 in row direction and a process module 4 in column direction;
the processing module 3 in the row direction comprises a sampling original image sub-module 31, a new pixel coordinate calculating sub-module 32 and a scaling signal calculating sub-module 33;
the column-direction processing module 4 comprises a sampling scaling row sub-module 41 and an updating row pixel coordinate sub-module 42;
the new image position calculating module 1 is used for calculating the coordinates and coverage of the actual display memory after the image is scaled and overturned;
the pixel module 2 of the dispatch row is used for filtering the pixel rows outside the column direction range and then dispatching the effective pixel rows;
the processing module 3 in the row direction is used for calculating a scaling factor to realize scaling of sampling pixels, then calculating coordinates of all scaled pixels in each row, and calculating a row number of the current scaled row corresponding to the scaled image;
the processing module 4 in the column direction is used for calculating the scaled line number samples, selecting the effective line number, and updating the coordinates of the pixels in the effective line.
Furthermore, the new image position calculating module 1 receives the scaling parameter, the turnover parameter, the original image width and height, the video memory range and the drawing coordinates,
calculating new image coordinates and coverage of the image in the actual video memory range after scaling and overturning,
the coverage, new map coordinates are sent to the dispatch row pixel module 2 via the TLM interface.
Further, the dispatch line pixel module 2 receives the original image, the scaling parameter, the flipping parameter, and calculates the coverage area and the new image coordinates sent by the new image position module 1,
the rows of pixels outside the column direction range are filtered, then the active rows of pixels are assigned,
the single-row original image, the scaling parameters, the overturning parameters, the coverage, the new image coordinates and the current row number are sent to a processing module 3 in the row direction through a TLM interface;
and meanwhile, the overturn parameters and the new graph coordinates are sent to the processing module 4 in the column direction through the TLM interface.
Further, the processing module 3 in the row direction receives a single row of original image, scaling parameters, flipping parameters, coverage, new image coordinates, current row number sent by the pixel module 2 in the assigned row,
a scaling factor is calculated based on the scaling parameters,
row pixel coordinates for each row are calculated based on the new map coordinates and the current row number,
a scaled post-line number is calculated based on the scaling factor and the current line number,
and then the scaled row pixels, row pixel coordinates and scaled row numbers are sent to the processing module 4 in the column direction through the TLM interface.
Further, the sub-module 31 for sampling the original image receives the single-line original image, the coverage, the scaling parameter, the flipping parameter sent by the pixel module 2 of the dispatch line,
the scaling factor is calculated and the scaling factor is calculated,
the original image line is then sampled and the sampled pixels are sent to the calculate new pixel coordinates sub-module 32 and the scaling factors are sent to the calculate scaling line number sub-module 33.
Further, the new pixel coordinate calculating sub-module 32 receives the scaling parameters, the inversion parameters, the coverage, the new image coordinates sent by the dispatch line pixel module 2, and samples the sampled pixels sent by the original image sub-module 31,
the coordinates of the new pixel are calculated and,
the new pixel coordinates and the sampled pixels are then sent to the processing module 4 in the column direction.
Further, the computation scaling signal sub-module 33 receives the inversion parameter sent by the dispatch line pixel module 2, the current line number, and the scaling factor sent by the sample original image sub-module 31,
the current zoom line is calculated to correspond to the zoom post-line number,
the scaled row number is then sent to the processing module 4 in the column direction.
Further, the processing module 4 in the column direction receives the scaled row pixels, the row pixel coordinates and the scaled row numbers sent by the processing module 3 in the row direction, assigns the flipping parameters and the new map coordinates sent by the row pixel module 2,
the effective number of rows is selected by scaling the post-row number samples,
and updating the coordinates of the pixels of the effective row, and finishing the rectangular scaling and overturning of the GPU pixels.
Further, the sampling scaling row sub-module 41 receives the scaled row number sent by the processing module 3 in the row direction,
the line number is effectively scaled by scaling the post-line number samples,
the valid zoom line number is sent to the update line pixel coordinates sub-module 42.
Further, the updated line pixel coordinate sub-module 42 receives the scaling flip 2 in the line direction, sends the flip parameters, the new map coordinates, the scaled line pixels, the line pixel coordinates, the scaled line number sent by the processing module 3 in the line direction, and the effective scaling line number sent by the sampling scaling line sub-module 41,
selecting a corresponding row of pixels by comparing the effective scaling row numbers,
and then updating the pixel coordinates of the effective row, and finishing the scaling and turning of the GPU pixel rectangle.
The invention has the beneficial effects that:
the invention realizes the function and the realization structure of the pixel rectangle scaling and overturning algorithm based on the TLM model, realizes the function defined by the glPixelZoom () function defined by the OpenGL API, solves the problem of insufficient performance of scaling and overturning the pixel rectangle by the GPU hardware, and effectively accelerates the RTL design and development.
Drawings
FIG. 1 is a block diagram of a hardware TLM micro-architecture of the pixel rectangle scaling flip algorithm of the present invention;
Detailed Description
The technical scheme of the invention is clearly and completely described below with reference to the accompanying drawings and the specific embodiments. It is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments, and that all other embodiments obtained by a person skilled in the art without making creative efforts based on the embodiments in the present invention are within the protection scope of the present invention.
The invention provides a TLM microstructure for a GPU pixel rectangular scaling and turning algorithm, which comprises a new graph position calculating module 1, a row pixel assigning module 2, a row direction processing module 3 and a column direction processing module 4;
the new image position calculating module 1 is used for calculating the coordinates and coverage of the actual display memory after the image is scaled and overturned;
the new image position calculating module 1 receives the scaling parameter, the turnover parameter, the width and height of the original image, the video memory range and the drawing coordinate,
calculating new image coordinates and coverage of the image in the actual video memory range after scaling and overturning,
the coverage range and the new map coordinates are sent to the dispatch row pixel module 2 through a TLM interface;
the pixel module 2 of the dispatch row is used for filtering the pixel rows outside the column direction range and then dispatching the effective pixel rows;
the dispatch row pixel module 2 receives the original image, the scaling parameters, the flipping parameters, and calculates the coverage, new map coordinates sent by the new map location module 1,
the rows of pixels outside the column direction range are filtered, then the active rows of pixels are assigned,
the single-row original image, the scaling parameters, the overturning parameters, the coverage, the new image coordinates and the current row number are sent to a processing module 3 in the row direction through a TLM interface;
simultaneously, the overturning parameters and the new graph coordinates are sent to a processing module 4 in the column direction through a TLM interface;
the processing module 3 in the row direction comprises a sampling original image sub-module 31, a new pixel coordinate calculating sub-module 32 and a scaling signal calculating sub-module 33;
the processing module 3 in the row direction is used for calculating a scaling factor to realize scaling of sampling pixels, then calculating coordinates of all scaled pixels in each row, and calculating a row number of the current scaled row corresponding to the scaled image;
the processing module 3 in the row direction receives the single-row original image, the scaling parameter, the flipping parameter, the coverage, the new image coordinates and the current row number sent by the pixel module 2 in the assigned row,
a scaling factor is calculated based on the scaling parameters,
row pixel coordinates for each row are calculated based on the new map coordinates and the current row number,
a scaled post-line number is calculated based on the scaling factor and the current line number,
and then the scaled row pixels, row pixel coordinates and scaled row numbers are sent to the processing module 4 in the column direction through the TLM interface. The sub-module 31 for sampling the original image receives the single line original image, the coverage, the scaling parameters, the flipping parameters sent by the sub-module 2 for assigning lines,
the scaling factor is calculated and the scaling factor is calculated,
the original image line is then sampled and the sampled pixels are sent to the calculate new pixel coordinates sub-module 32 and the scaling factors are sent to the calculate scaling line number sub-module 33.
The calculate new pixel coordinates sub-module 32 receives the scaling parameters, the flipping parameters, the coverage, the new map coordinates sent by the dispatch line pixel module 2, and samples the sampled pixels sent by the original image sub-module 31,
the coordinates of the new pixel are calculated and,
the new pixel coordinates and the sampled pixels are then sent to the processing module 4 in the column direction.
The calculate scaling parameters sent by the dispatch row pixel module 2, the current row number, and the scaling factor sent by the sample raw image sub-module 31 are received by the calculate scaling parameters sub-module 33,
the current zoom line is calculated to correspond to the zoom post-line number,
the scaled row number is then sent to the processing module 4 in the column direction.
The processing module 4 in the column direction is used for calculating the scaled line number samples, selecting the effective line number, and updating the coordinates of the pixels in the effective line.
The column-direction processing module 4 comprises a sampling scaling row sub-module 41 and an updating row pixel coordinate sub-module 42;
the column-direction processing module 4 receives the scaled row pixels, the row pixel coordinates, the scaled row numbers sent by the row-direction processing module 3, and assigns the flipping parameters and the new map coordinates sent by the row pixel module 2,
the effective number of rows is selected by scaling the post-row number samples,
and updating the coordinates of the pixels of the effective row, and finishing the rectangular scaling and overturning of the GPU pixels.
The sample scaling row sub-module 41 receives the scaled row number sent by the processing module 3 in the row direction,
the line number is effectively scaled by scaling the post-line number samples,
the valid zoom line number is sent to the update line pixel coordinates sub-module 42.
The update row pixel coordinate sub-module 42 receives the row-direction scaled flip 2 transmit flip parameters, the new map coordinates, the row-direction scaled post-row pixels, the row pixel coordinates, the scaled post-row number transmitted by the row-direction processing module 3, and the effective scaled row number transmitted by the sample scaled row sub-module 41,
selecting a corresponding row of pixels by comparing the effective scaling row numbers,
and then updating the pixel coordinates of the effective row, and finishing the scaling and turning of the GPU pixel rectangle.
Examples:
the present invention will be described in further detail with reference to fig. 1.
A TLM microstructure for GPU-pixel-oriented rectangular scaling and flipping algorithm, the structure comprising a module for calculating a new map location 1, a module for assigning row pixels 2, a processing module for row direction 3, and a processing module for column direction 4;
the new image position calculating module 1 is used for calculating the coordinates and coverage of the actual display memory after the image is scaled and overturned;
the new image position calculating module 1 receives the scaling parameter, the turnover parameter, the width and height of the original image, the video memory range and the drawing coordinate,
calculating new image coordinates and coverage of the image in the actual video memory range after scaling and overturning,
the coverage range and the new map coordinates are sent to the dispatch row pixel module 2 through a TLM interface;
the pixel module 2 of the dispatch row is used for filtering the pixel rows outside the column direction range and then dispatching the effective pixel rows;
the dispatch row pixel module 2 receives the original image, the scaling parameters, the flipping parameters, and calculates the coverage, new map coordinates sent by the new map location module 1,
the rows of pixels outside the column direction range are filtered, then the active rows of pixels are assigned,
the single-row original image, the scaling parameters, the overturning parameters, the coverage, the new image coordinates and the current row number are sent to a processing module 3 in the row direction through a TLM interface;
simultaneously, the overturning parameters and the new graph coordinates are sent to a processing module 4 in the column direction through a TLM interface;
the processing module 3 in the row direction comprises a sampling original image sub-module 31, a new pixel coordinate calculating sub-module 32 and a scaling signal calculating sub-module 33;
the processing module 3 in the row direction is used for calculating a scaling factor to realize scaling of sampling pixels, then calculating coordinates of all scaled pixels in each row, and calculating a row number of the current scaled row corresponding to the scaled image;
the processing module 3 in the row direction receives the single-row original image, the scaling parameter, the flipping parameter, the coverage, the new image coordinates and the current row number sent by the pixel module 2 in the assigned row,
a scaling factor is calculated based on the scaling parameters,
row pixel coordinates for each row are calculated based on the new map coordinates and the current row number,
a scaled post-line number is calculated based on the scaling factor and the current line number,
and then the scaled row pixels, row pixel coordinates and scaled row numbers are sent to the processing module 4 in the column direction through the TLM interface. The sub-module 31 for sampling the original image receives the single line original image, the coverage, the scaling parameters, the flipping parameters sent by the sub-module 2 for assigning lines,
the scaling factor is calculated and the scaling factor is calculated,
the original image line is then sampled and the sampled pixels are sent to the calculate new pixel coordinates sub-module 32 and the scaling factors are sent to the calculate scaling line number sub-module 33.
The calculate new pixel coordinates sub-module 32 receives the scaling parameters, the flipping parameters, the coverage, the new map coordinates sent by the dispatch line pixel module 2, and samples the sampled pixels sent by the original image sub-module 31,
the coordinates of the new pixel are calculated and,
the new pixel coordinates and the sampled pixels are then sent to the processing module 4 in the column direction.
The calculate scaling parameters sent by the dispatch row pixel module 2, the current row number, and the scaling factor sent by the sample raw image sub-module 31 are received by the calculate scaling parameters sub-module 33,
the current zoom line is calculated to correspond to the zoom post-line number,
the scaled row number is then sent to the processing module 4 in the column direction.
The processing module 4 in the column direction is used for calculating the scaled line number samples, selecting the effective line number, and updating the coordinates of the pixels in the effective line.
The column-direction processing module 4 comprises a sampling scaling row sub-module 41 and an updating row pixel coordinate sub-module 42;
the column-direction processing module 4 receives the scaled row pixels, the row pixel coordinates, the scaled row numbers sent by the row-direction processing module 3, and assigns the flipping parameters and the new map coordinates sent by the row pixel module 2,
the effective number of rows is selected by scaling the post-row number samples,
and updating the coordinates of the pixels of the effective row, and finishing the rectangular scaling and overturning of the GPU pixels.
The sample scaling row sub-module 41 receives the scaled row number sent by the processing module 3 in the row direction,
the line number is effectively scaled by scaling the post-line number samples,
the valid zoom line number is sent to the update line pixel coordinates sub-module 42.
The update row pixel coordinate sub-module 42 receives the row-direction scaled flip 2 transmit flip parameters, the new map coordinates, the row-direction scaled post-row pixels, the row pixel coordinates, the scaled post-row number transmitted by the row-direction processing module 3, and the effective scaled row number transmitted by the sample scaled row sub-module 41,
selecting a corresponding row of pixels by comparing the effective scaling row numbers,
and then updating the pixel coordinates of the effective row, and finishing the scaling and turning of the GPU pixel rectangle.
The GPU-oriented pixel rectangular scaling and turning algorithm based on the TLM microstructure comprises the following steps of:
step 1, calculating the position and the range of the image after zooming and overturning, and calculating the drawing range of a new image according to the original width and height of the image, the zooming parameter and the overturning parameter; judging the drawing initial position of the new image in the x direction and the y direction, if the initial position is smaller than the boundary, assigning 0 to the new image coordinate in the corresponding direction, otherwise, the new image coordinate is the same as the drawing coordinate; and finally, calculating the coverage range according to the new graph coordinates and the video memory range.
And step 2, assigning effective pixel rows, filtering the ineffective pixel rows according to the new drawing range, and then assigning the effective pixel rows.
And 3, scaling and turning in the row direction, firstly, calculating an x-direction scaling factor, and reversely pushing the sampling position of the original pixel row according to the scaling position and the scaling factor to sample the pixels. And then, calculating the coordinates of the pixels in the video memory according to the scaling parameters, the turning parameters, the new graph coordinates and the current line number. And finally, calculating the line number of the current zoom line corresponding to the zoomed image according to the current line number and the zoom parameter.
And 4, scaling and overturning in the column direction, sampling an effective scaling row through a scaling post-row number, and updating coordinates of the scaling post-row pixels according to scaling parameters, overturning parameters and row pixel coordinates. And finishing the rectangular scaling and overturning of the GPU pixels.
Claims (9)
1. The TLM microstructure for the GPU pixel rectangular scaling and turning algorithm is characterized in that: the structure comprises a new graph position calculating module 1, a row pixel distributing module 2, a row direction processing module 3 and a column direction processing module 4;
the processing module 3 in the row direction comprises a sampling original image sub-module 31, a new pixel coordinate calculating sub-module 32 and a scaling signal calculating sub-module 33;
the column-direction processing module 4 comprises a sampling scaling row sub-module 41 and an updating row pixel coordinate sub-module 42;
the new image position calculating module 1 is used for calculating the coordinates and coverage of the actual display memory after the image is scaled and overturned;
the pixel module 2 of the dispatch row is used for filtering the pixel rows outside the column direction range and then dispatching the effective pixel rows;
the processing module 3 in the row direction is used for calculating a scaling factor to realize scaling of sampling pixels, then calculating coordinates of all scaled pixels in each row, and calculating a row number of the current scaled row corresponding to the scaled image;
the processing module 4 in the column direction is used for calculating the scaled line number samples, selecting the effective line number, updating the coordinates of the pixels in the effective line,
the new image position calculating module 1 receives the scaling parameter, the turnover parameter, the width and height of the original image, the video memory range and the drawing coordinate,
calculating new image coordinates and coverage of the image in the actual video memory range after scaling and overturning,
the coverage, new map coordinates are sent to the dispatch row pixel module 2 via the TLM interface.
2. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the dispatch row pixel module 2 receives the original image, the scaling parameters, the flipping parameters, and calculates the coverage, new map coordinates sent by the new map location module 1,
the rows of pixels outside the column direction range are filtered, then the active rows of pixels are assigned,
the single-row original image, the scaling parameters, the overturning parameters, the coverage, the new image coordinates and the current row number are sent to a processing module 3 in the row direction through a TLM interface;
and meanwhile, the overturn parameters and the new graph coordinates are sent to the processing module 4 in the column direction through the TLM interface.
3. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the processing module 3 in the row direction receives the single-row original image, the scaling parameter, the flipping parameter, the coverage, the new image coordinates and the current row number sent by the pixel module 2 in the assigned row,
a scaling factor is calculated based on the scaling parameters,
row pixel coordinates for each row are calculated based on the new map coordinates and the current row number,
a scaled post-line number is calculated based on the scaling factor and the current line number,
and then the scaled row pixels, row pixel coordinates and scaled row numbers are sent to the processing module 4 in the column direction through the TLM interface.
4. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the sub-module 31 for sampling the original image receives the single line original image, the coverage, the scaling parameters, the flipping parameters sent by the sub-module 2 for assigning lines,
the scaling factor is calculated and the scaling factor is calculated,
the original image line is then sampled and the sampled pixels are sent to the calculate new pixel coordinates sub-module 32 and the scaling factors are sent to the calculate scaling line number sub-module 33.
5. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the calculate new pixel coordinates sub-module 32 receives the scaling parameters, the flipping parameters, the coverage, the new map coordinates sent by the dispatch line pixel module 2, and samples the sampled pixels sent by the original image sub-module 31,
the coordinates of the new pixel are calculated and,
the new pixel coordinates and the sampled pixels are then sent to the processing module 4 in the column direction.
6. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the calculate scaling parameters sent by the dispatch row pixel module 2, the current row number, and the scaling factor sent by the sample raw image sub-module 31 are received by the calculate scaling parameters sub-module 33,
the current zoom line is calculated to correspond to the zoom post-line number,
the scaled row number is then sent to the processing module 4 in the column direction.
7. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the column-direction processing module 4 receives the scaled row pixels, the row pixel coordinates, the scaled row numbers sent by the row-direction processing module 3, and assigns the flipping parameters and the new map coordinates sent by the row pixel module 2,
the effective number of rows is selected by scaling the post-row number samples,
and updating the coordinates of the pixels of the effective row, and finishing the rectangular scaling and overturning of the GPU pixels.
8. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the sample scaling row sub-module 41 receives the scaled row number sent by the processing module 3 in the row direction,
the line number is effectively scaled by scaling the post-line number samples,
the valid zoom line number is sent to the update line pixel coordinates sub-module 42.
9. The GPU-pixel-oriented rectangular scaling flip algorithm TLM microstructure of claim 1, wherein:
the update row pixel coordinate sub-module 42 receives the row-direction scaled flip 2 transmit flip parameters, the new map coordinates, the row-direction scaled post-row pixels, the row pixel coordinates, the scaled post-row number transmitted by the row-direction processing module 3, and the effective scaled row number transmitted by the sample scaled row sub-module 41,
selecting a corresponding row of pixels by comparing the effective scaling row numbers,
and then updating the pixel coordinates of the effective row, and finishing the scaling and turning of the GPU pixel rectangle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911125671.XA CN111028129B (en) | 2019-11-18 | 2019-11-18 | TLM microstructure for GPU pixel rectangular scaling and turning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911125671.XA CN111028129B (en) | 2019-11-18 | 2019-11-18 | TLM microstructure for GPU pixel rectangular scaling and turning algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111028129A CN111028129A (en) | 2020-04-17 |
| CN111028129B true CN111028129B (en) | 2023-09-15 |
Family
ID=70200395
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911125671.XA Active CN111028129B (en) | 2019-11-18 | 2019-11-18 | TLM microstructure for GPU pixel rectangular scaling and turning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111028129B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101950523A (en) * | 2010-09-21 | 2011-01-19 | 上海大学 | Adjustable rectangular window image scaling method and device |
| CN102800049A (en) * | 2012-08-08 | 2012-11-28 | 广东威创视讯科技股份有限公司 | Image reduction method and device |
| CN102831571A (en) * | 2011-07-08 | 2012-12-19 | 图芯芯片技术（上海）有限公司 | Design method of five-order filter for realizing graphic image resizing and rotation in one step in flow-line manner |
| CN103578077A (en) * | 2013-12-02 | 2014-02-12 | 广东威创视讯科技股份有限公司 | Image zooming method and related device |
| CN104363385A (en) * | 2014-10-29 | 2015-02-18 | 复旦大学 | Line-oriented hardware implementing method for image fusion |
| CN108154477A (en) * | 2017-12-26 | 2018-06-12 | 深圳市兴森快捷电路科技股份有限公司 | A kind of image rotating method based on FPGA |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1583031A1 (en) * | 2004-03-30 | 2005-10-05 | Dialog Semiconductor GmbH | Zoom algorithm |
| GB2499635B (en) * | 2012-02-23 | 2014-05-14 | Canon Kk | Image processing for projection on a projection screen |
-
2019
- 2019-11-18 CN CN201911125671.XA patent/CN111028129B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101950523A (en) * | 2010-09-21 | 2011-01-19 | 上海大学 | Adjustable rectangular window image scaling method and device |
| CN102831571A (en) * | 2011-07-08 | 2012-12-19 | 图芯芯片技术（上海）有限公司 | Design method of five-order filter for realizing graphic image resizing and rotation in one step in flow-line manner |
| CN102800049A (en) * | 2012-08-08 | 2012-11-28 | 广东威创视讯科技股份有限公司 | Image reduction method and device |
| CN103578077A (en) * | 2013-12-02 | 2014-02-12 | 广东威创视讯科技股份有限公司 | Image zooming method and related device |
| CN104363385A (en) * | 2014-10-29 | 2015-02-18 | 复旦大学 | Line-oriented hardware implementing method for image fusion |
| CN108154477A (en) * | 2017-12-26 | 2018-06-12 | 深圳市兴森快捷电路科技股份有限公司 | A kind of image rotating method based on FPGA |
Non-Patent Citations (5)
| Title |
|---|
| Faster algorithms for growing prioritized disks and rectangles;Heep-Kap Ahn;Computernal Geometry;第80卷;23-29 * |
| 覃方涛 等.CUDA并行技术与数字图像几何变换.计算机系统应用.2010,第19卷(第10期),第168-172，116页. * |
| 闫龙 等.《双目视觉测量系统相关技术研究》.济南：山东大学出版社,2017,第25-27页. * |
| 阳富民,赵宁,张杰.基于OpenGL的三维窗口裁剪、拾取算法研究.华中科技大学学报(自然科学版).2005,(第04期),第27-29，33页. * |
| 黄金凤 ; 陈小娥 ; .基于Seam Carving的图像自适应缩放算法研究.龙岩学院学报.2017,(第05期),第44-50页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111028129A (en) | 2020-04-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6323874B1 (en) | System and method for rendering an image | |
| CN101388157B (en) | Electronic map display method, apparatus and equipment | |
| WO2022089018A1 (en) | Method and apparatus for slicing three-dimensional vector data of three-dimensional vector map, and electronic device | |
| CN106296779A (en) | A kind of threedimensional model rendering indication method and system | |
| WO2022237811A1 (en) | Image processing method and apparatus, and device | |
| WO2019192165A1 (en) | Method and system for layered real-time graphics drawing and rendering | |
| CN111145286A (en) | Point nine picture mapping drawing method for edge calculation scene based on GPU | |
| CN106548500A (en) | A kind of two-dimension situation image processing method and device based on GPU | |
| CN103714213B (en) | Parallel drawing method for complicated geographic information | |
| KR102251444B1 (en) | Graphic processing unit, graphic processing system comprising the same, antialiasing method using the same | |
| WO2017095617A1 (en) | Writing graphics data from local memory to system memory | |
| CN106251291A (en) | Utilize OpenGL with OpenCL to cooperate and realize the method and system of image scaling | |
| CN111028129B (en) | TLM microstructure for GPU pixel rectangular scaling and turning algorithm | |
| CN101656037A (en) | Method for displaying large-format picture on small screen equipment and small screen equipment | |
| CN103310409B (en) | A kind of Tile-based renders the fast triangle block method of framework GPU | |
| CN111569418B (en) | Rendering method, device and medium for content to be output and electronic equipment | |
| CN106952216B (en) | Line detection method based on IMX6 graphics processor | |
| CN107688431B (en) | Man-machine interaction method based on radar positioning | |
| CN107301045B (en) | Window display method and device | |
| CN102364522B (en) | Large-scale weather volume data drawing method based on global division | |
| CN101799735B (en) | Primary handwriting hand input display method | |
| CN109741433B (en) | Triangle multidirectional parallel scanning method and structure based on Tile | |
| CN103675763B (en) | A kind of based on hardware-accelerated vector quantization VTS radar video rendering method | |
| CN112132871B (en) | Visual feature point tracking method and device based on feature optical flow information, storage medium and terminal | |
| CN111047498B (en) | GPU hardware copy buffer algorithm-oriented TLM microstructure |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |