Probability hypothesis density based variable target number video tracking algorithm

A probabilistic assumption density and video tracking technology, applied in the field of target tracking, can solve problems such as the inability to give target identity feature information, inability to track new targets, etc., and achieve the effect of improving performance

Inactive Publication Date: 2015-05-20
JIANGNAN UNIV
View PDF1 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method has two disadvantages: (1) Since the output of GM-PHD is a random set of state estimation, the target identity (track) feature information cannot be given; (2) Newborn targets with unknown positions cannot be tracked

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
  • Probability hypothesis density based variable target number video tracking algorithm
  • Probability hypothesis density based variable target number video tracking algorithm
  • Probability hypothesis density based variable target number video tracking algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0026] This embodiment uses the pedestrian detection video database "OneShopOneWait2front" to implement. Image 6 The setting values ​​of the main parameters in the embodiment are given. In this embodiment, a w×h rectangle is used to represent the target area, and the target state vector can be expressed as x k =(po x,k ,po y,k ,v x,k ,v y,k ,w,h) T , Where po k =(po x,k ,po y,k ) Is the center of the rectangle, v k =(v x,k ,v y,k ) Is the speed of the target center in the horizontal and vertical directions of the image, and an observation vector of the target is z k =(po k ,w,h) T . Assuming that the Gaussian parameter describing the posterior PHD at time k is Let the measured random set at k+1 be Among them, J k , They are the number of Gaussian elements of the posterior PHD at time k, the mean value, weight and corresponding covariance of the i-th Gaussian element. Suppose the freshman goal set at time k is Its Gaussian element parameter is { ω γ , k ...

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 probability hypothesis density based variable target number video tracking algorithm. The algorithm includes the following steps: adopting a motion detection method to realize quick effective segmentation of a motion target so as to generate target measurement; utilizing a motion detection result to design a newly generated target strength updating method so as to generate a newly generated target strength function; forecasting and updating PHD; trimming updated Gauss elements; extracting states and outputting estimated target numbers and states; adopting an auction track identification method to perform 'track-state estimation' association so as to identify a target track, and outputting the track with identity. By the algorithm, variable-number multiple targets can be effectively tracked, newly generated targets with tracking positions unknown can be tracked particularly, and the target track can be generated. By the algorithm, performance of GM-PHD filter is improved, robust estimation state can be obtained under the conditions of target number change and target crossing, and the target track can be identified.

Description

Technical field [0001] The invention relates to an algorithm in the field of target tracking, in particular to a video tracking algorithm based on the change of the target number of probability hypothesis density. Background technique [0002] Video target tracking is widely used in military and civilian applications, such as various defense systems in the military: battlefield surveillance systems, airborne fire control systems, etc., and civilian applications include automobiles and personal GPS navigation systems. The purpose of video target tracking is to establish correspondence between consecutive image frames based on target characteristics such as position, speed, color, shape, texture, etc. and target description models, and then obtain target status information such as target position and shape. General video target tracking methods can be divided into two categories: (1) Data-driven algorithms, which directly establish the similarity function between the target templat...

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): G06T7/20
Inventor 吴静静宋淑娟尤丽华王金华
Owner JIANGNAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap