Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

TLM microstructure for GPU hardware image processing convolution filtering system

An image processing and filtering system technology, applied in processor architecture/configuration, complex mathematical operations, etc., can solve problems such as difficult verification and debugging, large hardware logic scale, etc., and achieve the effect of speeding up RTL design and development

Active Publication Date: 2020-04-28
XIAN AVIATION COMPUTING TECH RES INST OF AVIATION IND CORP OF CHINA
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, when the GPU chip uses RTL to implement the above algorithm details, the hardware logic scale is huge, and it is difficult to verify and debug at the RTL stage.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • TLM microstructure for GPU hardware image processing convolution filtering system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0071] Below in conjunction with accompanying drawing the present invention is described in further detail, please refer to figure 1 .

[0072] A TLM microstructure for GPU hardware image processing convolution filtering system,

[0073] Including convolution kernel loading module 1, parameter initialization module 2, convolution data storage module 3, convolution filtering module 4 and pixel collection module 5;

[0074] The convolution kernel loading module 1 is used to write the data carried or copied by the load convolution kernel command into the convolution kernel according to the pixel type, data format, and internal format set in the command, and calculate the convolution kernel according to the convolution mode. The width and height of the core and half of the width and height;

[0075] The parameter initialization module 2 is used to initialize the initial address of the original pixel writing DDR, the number of rows of DDR stored data, the number of image processi...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to the technical field of computer hardware modeling, in particular to a TLM microstructure for a GPU hardware image processing convolution filtering system. The TLM microstructure oriented to the GPU hardware image processing convolution filtering algorithm comprises convolution kernel loading, convolution parameter initialization, convolution data storage, convolution filtering calculation and pixel collection after convolution. According to the method, the function and the implementation structure of the image processing convolution filtering algorithm based on the TLMmodel are realized, the problem of storage algorithm function verification for replacing the TLM microstructure with GPU hardware sub-textures is solved, and RTL design and development are effectively accelerated.

Description

technical field [0001] The invention relates to the technical field of computer hardware modeling, in particular to a TLM microstructure oriented to a GPU hardware image processing convolution filtering system. Background technique [0002] In the design and development of a graphics processor chip (hereinafter referred to as GPU), the correctness and efficiency of algorithms are important factors that determine the function and performance of the GPU. Convolution filtering is an important function in image processing. GPU hardware implementation of convolution function needs to comprehensively consider the convolution mode supported by OpenGL API, convolution kernel, convolved data, and DDR storage space limited by hardware architecture, original data storage and Problems such as the coordination of data addresses during convolution operations, and the unification of outputs in different convolution modes. However, when the GPU chip uses RTL to implement the above algorith...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06T1/20G06F17/15
CPCG06T1/20G06F17/15
Inventor 陈佳王绮卉姜丽云张少锋任向隆吴晓成
Owner XIAN AVIATION COMPUTING TECH RES INST OF AVIATION IND CORP OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products