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

Scale-direction self-adaptive Mean-shift tracking method aiming at video moving object

A technology of moving target and target scale, applied in image data processing, instrument, calculation and other directions, can solve the problems of neglect and achieve good results

Inactive Publication Date: 2011-07-06
NANJING UNIV
View PDF3 Cites 26 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Currently, target tracking methods based on SIFT features often only use SIFT feature descriptors and feature point location information, while ignoring the scale of feature points and the main direction information of feature points.

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
  • Scale-direction self-adaptive Mean-shift tracking method aiming at video moving object
  • Scale-direction self-adaptive Mean-shift tracking method aiming at video moving object
  • Scale-direction self-adaptive Mean-shift tracking method aiming at video moving object

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] The technical solution will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0043] The tracking method process of the present invention is as follows figure 1 As shown, the detailed implementation is as follows:

[0044] 1. Construct the target color histogram template q u , as shown in formula (1); extract the SIFT feature point set F of the template target, use a 128-dimensional feature point descriptor vector, and record the scale parameter σ of each feature point i mi and the main direction R mi . Set the initial value of the target position, scale, and direction to Y respectively 0 , S m and O m . Assume that the radius length of the target in the horizontal and vertical directions is h x 、h y , then the target scale is represented by the target transverse radius length, that is, S m = h x , in the follow-up operation first solve the target transverse radius length h x , the length of the longitudinal...

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 scale-direction self-adaptive Mean-shift tracking method aiming at a video moving object, comprising the following steps of: firstly, providing a prediction region for SIFT (Scale Invariant Feature Transform) feature detection by a Mean-shift algorithm in an image frame of a video sequence, and reducing a search range; secondly, tracking target scale variation by utilizing the acquired scale variation information of a current frame feature point corresponding to a reference frame feature point in the SIFT feature detection process; tracking target rotation movement by utilizing the acquired main direction variation information of the current frame feature point corresponding to the reference frame feature point in the SIFT feature detection process; regulating the bandwidth and the direction of a kernel function in the Mean-shift algorithm by utilizing the obtained target scale and target direction, and repeatedly executing the Mean-shift tracking; and updating a target template histogram of a reference frame and an SIFT feature point set when the target scale variation is over a certain threshold; therefore, the tracking with multiple degrees of freedom on positions, scales and directions of the target can be realized.

Description

technical field [0001] The invention belongs to the field of machine vision and video image processing, in particular to a scale-direction self-adaptive Mean-shift tracking method, which is mainly used in an intelligent monitoring system. Background technique [0002] Moving target tracking is a key step in intelligent video surveillance, and it is widely used in traffic monitoring, public security and other systems. Its main purpose is to locate the moving object of interest in each frame of the video sequence, and then refine the motion parameters of the object, such as centroid trajectory, direction change, scale change, etc. [0003] The Mean-shift algorithm was originally developed by the literature 【1】 Proposed, is a non-parametric density gradient ascent algorithm for finding extrema of the probability density function. In 2003 Comaniciu et al. 【2】 Applying it to target tracking, using the target color histogram as the feature, using the Bhattachary coefficient as ...

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 董蓉李勃陈启美江登表顾昊陈抒容丁文吴聪张潇
Owner NANJING UNIV
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