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

Video-based fast template matching GPU implementation method

A technology of template matching and implementation method, which is applied in special data processing applications, instruments, electrical digital data processing, etc. It can solve the problems of low template matching algorithm implementation speed and difficulty in meeting people's actual needs, so as to shorten operation time and improve implementation Efficiency, the effect of improving computing density

Active Publication Date: 2015-11-04
深圳市哈工交通电子有限公司
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0024] The purpose of the present invention is to provide a video-based fast template matching GPU implementation method to solve the problem that the implementation speed of the template matching algorithm in the prior art is low and it is difficult to meet the actual needs of people

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
  • Video-based fast template matching GPU implementation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0043] The video-based fast template matching GPU implementation method disclosed by the present invention is mainly aimed at the problem that the existing image template matching algorithm has low operating efficiency and is difficult to meet people's actual needs, and optimizes the algorithm. The invention does not change the basic principle of the existing template matching algorithm, but only adjusts the operation mode, can be directly applied to the existing equipment, has the characteristics of low cost, high efficiency, and simple and convenient implementation.

[0044] Such as figure 1 Shown, the specific implementation method of the present invention is as follows:

[0045] (1) Copy the image data to the GPU device. The operation of copying the image to the device and the image template matching are divided into two independent streams for simultaneous execution. One performs the image template matching calculation process operation of the current frame, and the other...

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 discloses a video-based fast template matching GPU implementation method which comprises the following steps: (1) copying image data to GPU equipment, wherein the operation of copying the images to the equipment and image template matching are executed at the same time as two independent streams, one of which executes a current-frame image template matching calculation process operation while the other of which executes a next-frame image copying operation; (2) storing the image template data in a shared memory, storing the image data in a texture memory, and randomly accessing by virtue of the texture memory for computation of match degree with the image template; (3) carrying out calculation at the same time by virtue of a multi-thread parallel calculation mode, wherein one thread calculates the template match degree quantity value of one position; and (4) determining the maximum value or the minimum value of the template match degree quantity value by virtue of a global atom and a shared atom so as to obtain a template matching result. According to the method disclosed by the invention, the template matching time can be obviously shortened and the practical value of a template matching algorithm can be improved.

Description

technical field [0001] The invention relates to a template matching method, in particular to a video-based fast template matching GPU implementation method. Background technique [0002] Template matching is a technique of finding the part of an image that best matches or is similar to another template image. Many video analysis application scenarios, such as vehicle and pedestrian tracking, have selectively added auxiliary functions of template matching to achieve the purpose of adapting to complex scenes. [0003] In order to determine the matching area, the template matching algorithm has to slide the template image and compare it with the original image. [0004] The image block slides one pixel at a time (from left to right, from top to bottom) at each position, and a metric calculation is performed to indicate how well it matches that position (or how similar the image block is to a specific region of the original image) degree), the position with the largest or smal...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/50
Inventor 曹泉余坚毅
Owner 深圳市哈工交通电子有限公司
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com