CN109979565B - Dosage calculation optimization method and device based on CUDA - Google Patents
Dosage calculation optimization method and device based on CUDA Download PDFInfo
- Publication number
- CN109979565B CN109979565B CN201910178033.8A CN201910178033A CN109979565B CN 109979565 B CN109979565 B CN 109979565B CN 201910178033 A CN201910178033 A CN 201910178033A CN 109979565 B CN109979565 B CN 109979565B
- Authority
- CN
- China
- Prior art keywords
- data
- dose
- radiotherapy
- calculation
- model
- 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
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/40—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
Landscapes
- Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Surgery (AREA)
- Urology & Nephrology (AREA)
- Engineering & Computer Science (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Image Generation (AREA)
Abstract
The invention discloses a dose calculation optimization method and a device based on CUDA (compute unified device architecture). By setting texture boundary attributes as periodic boundaries and then slipping data in a radiotherapy dose model to adapt to a calculation direction, the converted data can still keep the original data size, the problem that the original rotation method needs to occupy larger video memory space is avoided, the operation of data slipping is higher in speed and efficiency than the three-dimensional rotation operation using matrix multiplication operation in the original rotation method, meanwhile, the direction of the data after slipping is vertical to the calculation direction, reading conflict and merged access can be avoided when GPU (graphics processing Unit) is used for calculation, and therefore the efficiency and the speed of GPU calculation are greatly improved.
Description
Technical Field
The invention belongs to the field of radiotherapy dose calculation, and particularly relates to a dose calculation optimization method and device based on CUDA.
Background
Radiation dose distribution needs to be calculated in radiotherapy plan verification, and algorithms for comparing mainstream include a pencil beam (pencil beam) algorithm, a CCCS algorithm and the like, wherein the CCCS algorithm is a more common algorithm. The CCCS algorithm comprises a large number of calculations which can be performed in parallel and small calculations can be strengthened, and the steps are very suitable for calling a GPU (graphics processing Unit) to perform calculation by using CUDA (compute unified device architecture) software. To fully exert the performance of GPU computation, attention needs to be paid to the manner of reading data in global memory (global memory) and shared memory (shared memory). In particular, it is necessary to avoid the occurrence of read conflicts (reads) in the shared memory. By read conflict, it is meant that two or more threads read any byte in different 32-bit words in the same group (bank) at the same time. In addition, merging access (coordinated access) is achieved in the global memory reading as much as possible, namely, the physical addresses of the data processed by each thread are continuous, and the method is also beneficial to improving the calculation efficiency of the GPU.
When the CCCS algorithm is used for calculating the total dose, iterative calculation needs to be carried out on the passed voxel points along each direction of a radiotherapy dose model, and the results in each direction are accumulated to obtain the final dose. However, in a computer, the data arrangement direction of the radiotherapy dose model is often not perpendicular to the actual dose calculation direction, and in this case, the dose calculation may cause the read conflict and the condition of being unable to combine the accesses, thereby greatly reducing the calculation efficiency of the GPU.
To solve this problem, in the prior art, a method of rotating data in a video memory is often used, specifically, a three-dimensional cube composed of data is rotated three-dimensionally, the data is rotated to be right opposite to a calculation direction, and then calculation is performed. This method requires an additional piece of video memory to be allocated to store the rotated data, and since the cube shape is irregular after the rotation, the area for storing the data is usually larger than that before the rotation. The rotation requires a certain amount of calculation, which affects the efficiency of the algorithm.
As shown in fig. 1, the two-dimensional case is illustrated, where a rectangle is the original data arrangement and an arrow line is the calculated direction. The prior art rotates the data block of the left image into the right image through the rotation, so that the calculation direction becomes the vertical direction. However, as can be seen from the right drawing, a larger video memory space is required to store the original data after the rotation, and the new data arrangement mode has resource waste at the four corners of the rectangle. This situation is more pronounced in the three-dimensional case.
Disclosure of Invention
In order to overcome the technical defects, the invention provides a dosage calculation optimization method and device based on the CUDA, which can avoid the problem that the original rotation method needs to occupy larger video memory space, has higher speed and higher efficiency on data slippage operation than the three-dimensional rotation operation using matrix multiplication operation in the original rotation method, and greatly improves the efficiency and speed of GPU calculation.
In order to solve the problems, the invention is realized according to the following technical scheme:
a CUDA-based dose calculation optimization method comprises the following steps:
s1, setting a texture boundary attribute of CUDA software as a periodic boundary;
s2, calculating an included angle between the dose calculation direction and the data arrangement direction of the radiotherapy dose model; calculating the distance of each layer of data in the radiotherapy dose model which needs to move in different directions according to the included angle;
s3, slipping each layer of data of the radiotherapy dose model according to the distance to enable the data arrangement direction of the radiotherapy dose model to be vertical to the dose calculation direction;
and S4, accumulating and calculating the data of the radiotherapy dose model by using a CCCS algorithm to obtain the total dose of the radiotherapy dose model.
Further, the method also comprises the following steps:
and S5, performing inverse transformation on the radiotherapy dose model according to the direction and the distance, and returning the slid data to restore the radiotherapy dose model to the original data arrangement mode.
Further, the step S2 includes:
s21, establishing a standard three-dimensional coordinate system by taking the dose calculation direction as a Z axis;
s22, respectively calculating included angles between the data arrangement direction and an X axis and an Y axis of the standard three-dimensional coordinate system;
and S23, calculating the X-direction moving distance and the Y-direction moving distance of each layer of data of the radiotherapy dose model through a trigonometric function formula according to the included angle.
Further, the step S3 includes:
s31, determining the corresponding relation between the coordinates after the data in the radiotherapy dose model slips and the original coordinates according to the X-direction moving distance and the Y-direction moving distance;
and S32, carrying out coordinate change on all data in the radiotherapy dose model according to the corresponding relation so as to complete slippage, and enabling the data arrangement direction of the radiotherapy dose model to be vertical to the dose calculation direction.
Further, before step S2, the method further includes: placing the data of the radiotherapy dose model into a texture memory for temporary storage;
before step S4, the method further includes: and extracting the data of the radiotherapy dose model from the texture memory.
The invention also correspondingly discloses a dosage calculation optimization device based on CUDA, which comprises:
the boundary setting module is used for setting the texture boundary attribute of the CUDA software as a periodic boundary;
the distance calculation module is used for calculating an included angle between the dose calculation direction and the data arrangement direction of the radiotherapy dose model; calculating the distance of each layer of data in the radiotherapy dose model which needs to move in different directions according to the included angle;
the data slippage module is used for slipping data of each layer of the radiotherapy dose model according to the distance, so that the data arrangement direction of the radiotherapy dose model is perpendicular to the dose calculation direction;
and the accumulation calculation module is used for performing accumulation calculation on the data of the radiotherapy dose model by using a CCCS algorithm to obtain the total dose of the radiotherapy dose model.
Further, the device further comprises:
and the inverse transformation module is used for carrying out inverse transformation on the radiotherapy dosage model according to the direction and the distance, and returning the data after sliding so as to restore the radiotherapy dosage model to the original data arrangement mode.
Further, the distance calculation module includes:
a coordinate system establishing unit for establishing a standard three-dimensional coordinate system by taking the dose calculation direction as a Z axis;
the included angle calculation unit is used for calculating included angles between the data arrangement direction and an X axis and a Y axis of the standard three-dimensional coordinate system respectively;
and the moving distance calculating unit is used for calculating the moving distance in the X direction and the moving distance in the Y direction of each layer of data of the radiotherapy dose model according to the included angle and a trigonometric function formula.
Further, the data slipping module comprises:
the coordinate corresponding unit is used for determining the corresponding relation between the coordinates after the data in the radiotherapy dose model slides and the original coordinates according to the X-direction moving distance and the Y-direction moving distance;
and the coordinate mapping unit is used for carrying out coordinate modification on all data in the radiotherapy dose model according to the corresponding relation so as to complete slippage, so that the data arrangement direction of the radiotherapy dose model is perpendicular to the dose calculation direction.
Further, the device further comprises:
and the data migration module is used for temporarily storing the data of the radiotherapy dose model in a texture memory or extracting the data of the radiotherapy dose model from the texture memory.
Compared with the prior art, the invention has the beneficial effects that:
the invention discloses a dose calculation optimization method and a device based on CUDA (compute unified device architecture). By setting texture boundary attributes as periodic boundaries and then slipping data in a radiotherapy dose model to adapt to a calculation direction, the converted data can still keep the original data size, the problem that the original rotation method needs to occupy larger video memory space is avoided, the operation of data slipping is higher in speed and efficiency than the three-dimensional rotation operation using matrix multiplication operation in the original rotation method, meanwhile, the direction of the data after slipping is vertical to the calculation direction, reading conflict and merged access can be avoided when GPU (graphics processing Unit) is used for calculation, and therefore the efficiency and the speed of GPU calculation are greatly improved.
Drawings
FIG. 1 is a schematic data processing diagram of the rotation method in an embodiment of the present invention;
FIG. 2 is a schematic diagram showing the steps of the CUDA-based dose calculation optimization method according to embodiment 1 of the present invention;
FIG. 3 is a schematic structural diagram of a CUDA-based dose calculation optimization device in embodiment 2 of the present invention;
fig. 4-10 are three-dimensional schematic diagrams of data processing as described in the detailed description of the invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Example 1
As shown in fig. 2, the present embodiment discloses a dosage calculation optimization method based on CUDA, including:
s1, setting a texture boundary attribute of CUDA software as a periodic boundary;
s2, calculating an included angle between the dose calculation direction and the data arrangement direction of the radiotherapy dose model; calculating the distance of each layer of data in the radiotherapy dose model which needs to move in different directions according to the included angle;
s3, slipping each layer of data of the radiotherapy dose model according to the distance to enable the data arrangement direction of the radiotherapy dose model to be vertical to the dose calculation direction;
and S4, accumulating and calculating the data of the radiotherapy dose model by using a CCCS algorithm to obtain the total dose of the radiotherapy dose model.
According to the optimization method disclosed by the invention, the texture boundary attribute is set as the periodic boundary, and then the data in the radiotherapy dose model is slipped to adapt to the calculation direction, so that the converted data still keeps the original data size, the problem that the original rotation method needs to occupy larger video memory space is avoided, the operation of data slipping is higher in speed and higher in efficiency than the three-dimensional rotation operation using matrix multiplication operation in the original rotation method, meanwhile, the direction of the slipped data is vertical to the calculation direction, reading conflict can be avoided and merged access can be achieved when the GPU is used for calculation, and therefore, the efficiency and the speed of the GPU calculation are greatly improved.
Further, the method also comprises the following steps:
and S5, performing inverse transformation on the radiotherapy dose model according to the direction and the distance, and returning the slid data to restore the radiotherapy dose model to the original data arrangement mode.
The purpose of this step is that the radiotherapy dose model should recover the original data arrangement mode after dose calculation, so as to facilitate the subsequent other processing operations.
Specifically, step S2 includes:
s21, establishing a standard three-dimensional coordinate system by taking the dose calculation direction as a Z axis;
s22, respectively calculating included angles between the data arrangement direction and the X axis and Y axis of the standard three-dimensional coordinate system;
and S23, calculating the X-direction moving distance and the Y-direction moving distance of each layer of data of the radiotherapy dose model through a trigonometric function formula according to the included angle.
Specifically, step S3 includes:
s31, determining the corresponding relation between the coordinates after the data in the radiotherapy dose model slips and the original coordinates according to the X-direction moving distance and the Y-direction moving distance;
and S32, carrying out coordinate change on all data in the radiotherapy dose model according to the corresponding relation so as to complete slippage, and enabling the data arrangement direction of the radiotherapy dose model to be vertical to the dose calculation direction.
Specifically, the key codes actually implemented in the CUDA software in step S3 are as follows:
float u＝(xOut+dx*zOut+0.5f)/(float)Xcount；
float v＝(yOut+dy*zOut+0.5f)/(float)Ycount；
float w＝(zOut+0.5f)/(float)Zcount；
newMat[xOut+Xcount*(yOut+Ycount*zOut)]＝tex3D<float>(texObj,u,v,w)；
in the above code, dx, dy are the increments of the distance each layer needs to be slipped along the z-direction in the x and y directions, respectively. And dx × zOut calculates the distance of the zOut layer required to slide in the X direction, wherein xCount, yCount and zCount are the sizes of the three-dimensional array data of the radiotherapy dose model in the X, Y and Z directions respectively.
Further, step S2 is preceded by: placing the data of the radiotherapy dose model into a texture memory for temporary storage;
step S4 also includes: and extracting data of the radiotherapy dose model from the texture memory.
Data of the radiotherapy dose model is put into the texture memory and extracted when the data is used, and the GPU is enabled to process memory access of a local space more efficiently by means of high-efficiency reading performance of the texture memory, and calculation efficiency is improved.
Example 2
As shown in fig. 3, the present embodiment corresponds to the optimization method of embodiment 1, and discloses a dosage calculation optimization device based on CUDA, which includes:
the boundary setting module 1 is used for setting the texture boundary attribute of the CUDA software as a periodic boundary;
the distance calculation module 2 is used for calculating an included angle between the dose calculation direction and the data arrangement direction of the radiotherapy dose model; calculating the distance of each layer of data in the radiotherapy dose model which needs to move in different directions according to the included angle;
the data slipping module 3 is used for slipping data of each layer of the radiotherapy dose model according to the distance, so that the data arrangement direction of the radiotherapy dose model is perpendicular to the dose calculation direction;
the accumulation calculation module 4 is used for carrying out accumulation calculation on the data of the radiotherapy dose model by using a CCCS algorithm to obtain the total dose of the radiotherapy dose model;
and the inverse transformation module 5 is used for performing inverse transformation on the radiotherapy dosage model according to the direction and the distance, and returning the data after sliding so as to restore the radiotherapy dosage model to the original data arrangement mode.
And the data migration module 6 is used for temporarily storing the data of the radiotherapy dose model in the texture memory or extracting the data of the radiotherapy dose model from the texture memory.
Further, the distance calculation module 2 includes:
a coordinate system establishing unit for establishing a standard three-dimensional coordinate system by taking the dose calculation direction as a Z axis;
the included angle calculating unit is used for respectively calculating the included angles between the data arrangement direction and the X axis and the Y axis of the standard three-dimensional coordinate system;
and the moving distance calculating unit is used for calculating the moving distance in the X direction and the moving distance in the Y direction of each layer of data of the radiotherapy dosage model through a trigonometric function formula according to the included angle.
Further, the data slipping module 3 includes:
the coordinate corresponding unit is used for determining the corresponding relation between the coordinates after the data in the radiotherapy dose model slips and the original coordinates according to the X-direction moving distance and the Y-direction moving distance;
and the coordinate mapping unit is used for carrying out coordinate modification on all data in the radiotherapy dose model according to the corresponding relation so as to complete slippage, so that the data arrangement direction of the radiotherapy dose model is vertical to the dose calculation direction.
The CUDA-based dose calculation optimization device disclosed in this embodiment corresponds to the CUDA-based dose calculation optimization method disclosed in embodiment 1, and the technical effects thereof are also similar, and are not described herein again.
Next, the data processing performed by the method or the apparatus disclosed in the above embodiments is visually shown by a three-dimensional schematic diagram:
as shown in fig. 4, the rectangular parallelepiped is data of the radiotherapy dose model, and four straight lines intersecting with the rectangular parallelepiped are dose calculation directions, and with the method or the apparatus disclosed above, as shown in fig. 5 to 10, data in the radiotherapy dose model different from the dose calculation directions are slipped to corresponding positions, so that the data arrangement direction (i.e., parallel arrangement in each layer) in the radiotherapy dose model is perpendicular to the dose calculation direction, thereby realizing merged access and accelerating the calculation efficiency of the GPU.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, so that any modification, equivalent change and modification made to the above embodiment according to the technical spirit of the present invention will still fall within the scope of the technical solution of the present invention.
Claims (10)
1. A CUDA-based dose calculation optimization method is characterized by comprising the following steps:
s1, setting a texture boundary attribute of CUDA software as a periodic boundary;
s2, calculating an included angle between the dose calculation direction and the data arrangement direction of the radiotherapy dose model; calculating the distance of each layer of data in the radiotherapy dose model which needs to move in different directions according to the included angle;
s3, slipping each layer of data of the radiotherapy dose model according to the distance to enable the data arrangement direction of the radiotherapy dose model to be vertical to the dose calculation direction;
and S4, accumulating and calculating the data of the radiotherapy dose model by using a CCCS algorithm to obtain the total dose of the radiotherapy dose model.
2. The CUDA-based dose calculation optimization method of claim 1, further comprising:
and S5, performing inverse transformation on the radiotherapy dose model according to the direction and the distance, and returning the slid data to restore the radiotherapy dose model to the original data arrangement mode.
3. The CUDA-based dose calculation optimization method of claim 1, wherein the step S2 comprises:
s21, establishing a standard three-dimensional coordinate system by taking the dose calculation direction as a Z axis;
s22, respectively calculating included angles between the data arrangement direction and an X axis and an Y axis of the standard three-dimensional coordinate system;
and S23, calculating the X-direction moving distance and the Y-direction moving distance of each layer of data of the radiotherapy dose model through a trigonometric function formula according to the included angle.
4. The CUDA-based dose calculation optimization method of claim 3, wherein the step S3 comprises:
s31, determining the corresponding relation between the coordinates after the data in the radiotherapy dose model slips and the original coordinates according to the X-direction moving distance and the Y-direction moving distance;
and S32, carrying out coordinate change on all data in the radiotherapy dose model according to the corresponding relation so as to complete slippage, and enabling the data arrangement direction of the radiotherapy dose model to be vertical to the dose calculation direction.
5. The CUDA-based dose calculation optimization method of claim 3, wherein the step S2 is preceded by: placing the data of the radiotherapy dose model into a texture memory for temporary storage;
before the step S4, the method further includes: and extracting the data of the radiotherapy dose model from the texture memory.
6. A CUDA-based dose calculation optimization device, comprising:
the boundary setting module is used for setting the texture boundary attribute of the CUDA software as a periodic boundary;
the distance calculation module is used for calculating an included angle between the dose calculation direction and the data arrangement direction of the radiotherapy dose model; calculating the distance of each layer of data in the radiotherapy dose model which needs to move in different directions according to the included angle;
the data slippage module is used for slipping each layer of data of the radiotherapy dose model according to the distance, so that the data arrangement direction of the radiotherapy dose model is vertical to the dose calculation direction;
and the accumulation calculation module is used for performing accumulation calculation on the data of the radiotherapy dose model by using a CCCS algorithm to obtain the total dose of the radiotherapy dose model.
7. The CUDA-based dose calculation optimization device of claim 6, further comprising:
and the inverse transformation module is used for performing inverse transformation on the radiotherapy dose model according to the direction and the distance, and returning the data after sliding so as to restore the radiotherapy dose model to the original data arrangement mode.
8. The CUDA-based dose calculation optimization device of claim 6, wherein the distance calculation module comprises:
a coordinate system establishing unit for establishing a standard three-dimensional coordinate system by taking the dose calculation direction as a Z axis;
the included angle calculation unit is used for calculating included angles between the data arrangement direction and an X axis and a Y axis of the standard three-dimensional coordinate system respectively;
and the moving distance calculating unit is used for calculating the moving distance in the X direction and the moving distance in the Y direction of each layer of data of the radiotherapy dose model according to the included angle and a trigonometric function formula.
9. The CUDA-based dose calculation optimization device of claim 8, wherein the data slip module comprises:
the coordinate corresponding unit is used for determining the corresponding relation between the coordinates after the data in the radiotherapy dose model slides and the original coordinates according to the X-direction moving distance and the Y-direction moving distance;
and the coordinate mapping unit is used for carrying out coordinate modification on all data in the radiotherapy dose model according to the corresponding relation so as to complete slippage, so that the data arrangement direction of the radiotherapy dose model is perpendicular to the dose calculation direction.
10. The CUDA-based dose calculation optimization device of claim 6, further comprising:
and the data migration module is used for temporarily storing the data of the radiotherapy dosage model in a texture memory or extracting the data of the radiotherapy dosage model from the texture memory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910178033.8A CN109979565B (en) | 2019-03-08 | 2019-03-08 | Dosage calculation optimization method and device based on CUDA |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910178033.8A CN109979565B (en) | 2019-03-08 | 2019-03-08 | Dosage calculation optimization method and device based on CUDA |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109979565A CN109979565A (en) | 2019-07-05 |
| CN109979565B true CN109979565B (en) | 2022-11-15 |
Family
ID=67078331
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910178033.8A Active CN109979565B (en) | 2019-03-08 | 2019-03-08 | Dosage calculation optimization method and device based on CUDA |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109979565B (en) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101664583B (en) * | 2009-09-09 | 2012-05-09 | 深圳市海博科技有限公司 | Dosage calculation optimization method and system based on CUDA |
| JP2018049335A (en) * | 2016-09-20 | 2018-03-29 | 富士ゼロックス株式会社 | Data creation device, molding device, molding method and program |
| CN106943679B (en) * | 2017-04-24 | 2019-04-16 | 安徽慧软科技有限公司 | Photon and electron dose calculate method under magnetic field based on GPU Monte carlo algorithm |
-
2019
- 2019-03-08 CN CN201910178033.8A patent/CN109979565B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109979565A (en) | 2019-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110276820B (en) | GIS model optimization method and system based on LOD organization and scheduling method | |
| CN107358238A (en) | A kind of method and system for extracting image feature information | |
| DE102020112826A1 (en) | PROCESS FOR EFFICIENT PERFORMANCE OF DATA REDUCTION IN PARALLEL PROCESSING UNITS | |
| WO2020125062A1 (en) | Image fusion method and related device | |
| CN112085033A (en) | Template matching method and device, electronic equipment and storage medium | |
| CN102810133B (en) | Ray querying method in online game and scene server | |
| CN112799599A (en) | Data storage method, computing core, chip and electronic equipment | |
| CN110648356A (en) | Multi-target tracking processing optimization method based on visual digital signal processing | |
| CN109979565B (en) | Dosage calculation optimization method and device based on CUDA | |
| CN106780415A (en) | A kind of statistics with histogram circuit and multimedia processing system | |
| CN113791425A (en) | Radar P display interface generation method and device, computer equipment and storage medium | |
| US10657620B2 (en) | Pooling method, device, and system, computer-readable storage medium | |
| CN108986212B (en) | Three-dimensional virtual terrain LOD model generation method based on crack elimination | |
| Kim et al. | Optimizing seam carving on multi-GPU systems for real-time content-aware image resizing | |
| US20100185425A1 (en) | Performing Molecular Dynamics Simulation on a Multiprocessor System | |
| JP2023092446A (en) | Cargo counting method and apparatus, computer apparatus, and storage medium | |
| CN107622037A (en) | The method and apparatus that a kind of Matrix Multiplication for improving graphics processing unit calculates performance | |
| US9805497B2 (en) | Collision-culling of lines over polygons | |
| CN114372928A (en) | Data processing method and device and electronic equipment | |
| EP1532582A1 (en) | Method and apparatus for image data processing using image strips and circular addressing arrangement | |
| CN112817972A (en) | Data storage method, data query method, data storage device, data query device and electronic equipment | |
| CN112686992A (en) | Geometric figure view frustum realization method and device for OCC tree in smart city and storage medium | |
| CN112288649A (en) | Image correction method and device for cylindrical object perspective imaging distortion | |
| CN109493406A (en) | Quick percentage is close to soft shadows method for drafting | |
| CN111612685B (en) | GPU dynamic self-adaptive acceleration method for remote sensing image |
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 |