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

Particle filter target tracking method and device based on pipeline parallel processing technology

A parallel processing and particle filter technology, applied in image data processing, instrumentation, computing, etc., to achieve the effects of increasing parallelism, compressing weight update, improving execution efficiency and real-time performance

Inactive Publication Date: 2016-01-20
NANJING NORMAL UNIVERSITY
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method can greatly improve the execution efficiency of the particle filter algorithm, improve real-time performance, and solve the bottleneck problem of memory access. On the other hand, it also proposes a scalable general-purpose parallel particle filter system framework to meet the requirements of algorithm improvement and upgrading

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
  • Particle filter target tracking method and device based on pipeline parallel processing technology
  • Particle filter target tracking method and device based on pipeline parallel processing technology
  • Particle filter target tracking method and device based on pipeline parallel processing technology

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] In order to better understand the technical solution of the present invention, in conjunction with the accompanying drawings, a further detailed description is as follows:

[0038] Here, the processing process is divided into three major steps: input preprocessing, parallel calculation of particle weights, and output processing. These three steps are processed by figure 2 The three major modules in the system block diagram are "input processing unit", "co-processing unit array" and "output processing unit". However, it should be noted that these three modules are in the pipeline parallel processing mode. While the "input processing unit" is processing the image of the kth frame, the latter two modules are processing the information of the k-1th frame image.

[0039] Step 1: Perform input preprocessing on the image, this part of the processing includes figure 1 The first 3 steps in the flowchart, meanwhile, this part of the processing work will be done by figure 2 T...

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 method and a device for particle filter target tracking based on a pipeline parallel processing technique. The method comprises the following steps: forecasting particle sets of this frame image, carrying out the distribution of data to be processed according to the particles, calculating particle weight by using the pipeline parallel processing technique, carrying out state renewing, figuring out the state estimation of a target in a new frame, and therefore tracking out the new position of the target, and meanwhile carrying out resampling to particles, avoiding particle degeneracy. The method uses the pipeline parallel processing technique, distributes each processing step to each link of the parallel pipeline and carries out parallel processing, and the execution efficiency and instantaneity of the particle filter algorithmic are greatly improved. Meanwhile the problems of a particle weight calculation bottleneck and a memorizer accessing bottleneck of a traditional particle algorithmic are solved. In addition, an extensible general parallel particle filter system framework is provided, and is beneficial for carrying out the expansion and upgrading base on the system framework.

Description

technical field [0001] The invention relates to a particle filter target tracking method and device based on pipeline parallel processing technology, which can significantly improve the execution efficiency and real-time performance of the particle filter target tracking algorithm, and belongs to the technical field of computer vision. Background technique [0002] Object tracking in image sequences is a research hotspot in the field of computer vision, which provides an important basis for video surveillance, precision guidance, robot navigation, video compression, etc. [0003] Particle filter (ParticleFilter) is an effective non-rigid target tracking algorithm proposed in recent years. It is essentially a method based on Monte Carlo simulation. It first obtains the reference color histogram of the target, and then uses a set of weighted random samples (particles) in the state space to approximate the probability density function of the state variable. Each The sample rep...

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 Patents(China)
IPC IPC(8): G06T7/20
Inventor 宁晨曾毓敏
Owner NANJING NORMAL UNIVERSITY
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