Supercharge Your Innovation With Domain-Expert AI Agents!

Multi-target video tracking method based on box particle PHD (Probability Hypothesis Density) filtering

A technology of video tracking and binning particles, which is applied in the field of information processing and can solve problems such as high computational complexity and low computational efficiency

Inactive Publication Date: 2016-10-12
XIDIAN UNIV
View PDF1 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the defective that above-mentioned prior art exists, has proposed a kind of multi-target video tracking method based on box particle PHD, on the basis of utilizing frame difference method to obtain multi-target preliminary position information and setting up multi-target motion model, Through the box particle PHD filtering method, the precise position information of multiple targets is obtained, which is used to solve the technical problems of high computational complexity and low operational efficiency in the existing particle PHD multi-target video tracking method

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
  • Multi-target video tracking method based on box particle PHD (Probability Hypothesis Density) filtering
  • Multi-target video tracking method based on box particle PHD (Probability Hypothesis Density) filtering
  • Multi-target video tracking method based on box particle PHD (Probability Hypothesis Density) filtering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments.

[0032] refer to figure 1 , the present invention comprises the following steps:

[0033] Step 1 target position initialization

[0034] Obtain the initial position of the target manually or by using the frame difference method. According to the appearance position and disappearance position of each target, divide it by the number of frames to roughly obtain the target's motion speed and judge the target's motion state. Generate N box particles centered on the initial position and obey the Gaussian distribution N(m,∑) box particles, and assign the same weight to each box particle m is the initial position matrix of the target, including position coordinates and velocity, and Σ is the covariance matrix.

[0035] Step 2 Build a motion model

[0036] Using the estimated target motion velocity, set the state transition function f k|k-1 , in the...

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 puts forward a multi-target video tracking method based on box particle PHD (Probability Hypothesis Density) filtering, and is used for solving the technical problems of high calculation complexity and low operation efficiency in a traditional particle PHD multi-target video tracking method. The multi-target video tracking method comprises the following implementation steps: carrying out target position initialization; establishing a motion model; carrying out earlier stage processing on a video target; initializing a box particle set; on the basis of box particle PHD prediction, obtaining a prediction box particle set; on the basis of box particle PHD update, obtaining a contraction box particle and a box particle update weight; carrying out summation on the box particle update weight to obtain a target number; adopting a random sub-division resampling method to resample the contraction box particle; clustering the sampled box particle set to obtain a target state set; and finally, importing the target state set into a video, and outputting. The multi-target video tracking method has the characteristics of being low in calculation complexity and high in operation efficiency, and can be used in the field of pedestrian vehicle monitoring in traffic flow and security management.

Description

technical field [0001] The invention belongs to the technical field of information processing, and relates to a multi-target video tracking method, in particular to a multi-target video tracking method based on box particle PHD filtering. By combining box particle filtering and probability hypothesis density PHD filtering, multiple Efficient tracking of targets can be used to monitor pedestrians and vehicles in traffic flow and security management. Background technique [0002] The target tracking technology is to process the obtained measurement information, and then estimate the target state, so as to achieve the tracking effect. In a multi-target environment, the number of targets is an important unknown as is the status of the targets. How to track multiple targets with varying numbers of targets in a clutter environment has always been a complex research topic in the academic and engineering fields. [0003] For multi-target video tracking methods, they can be roughly...

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
CPCG06T2207/10016
Inventor 宋骊平程慧姬红兵宋文斌
Owner XIDIAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com