Image tracking method and system thereof

A technology of image tracking and tracking area, which is applied in the field of image tracking, can solve the problems of large impact of tracking algorithm, large amount of calculation, and affecting real-time tracking, so as to avoid tracking failure, ensure tracking effect and improve real-time tracking

Active Publication Date: 2009-03-25
中星智能系统技术有限公司
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The Meanshift algorithm adopts an iterative strategy based on the gradient descent of the matching degree, and can only find the local optimal position of the matching degree, so that the tracking algorithm is greatly affected by the nearby interference target.
However, the target tracking algorithm based on global search needs to exhaustively search every window where the target may appear, and needs to calculate the histogram for each search window, and match the calculated histogram with the standard histogram of the target. operation, and the calculation of the histogram is relatively large, which makes the tracking speed slower and affects the real-time performance of the tracking

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
  • Image tracking method and system thereof
  • Image tracking method and system thereof
  • Image tracking method and system thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0078] figure 1 It is a flow chart of the image tracking method in Embodiment 1 of the present invention. Such as figure 1 As shown, the process includes the following steps:

[0079] Step 101, use the classifier to describe the confidence level of each search window where the target may appear in the entire tracking area of ​​the current image, if there is a search window whose confidence level meets the tracking requirements, then perform step 102; if there is no confidence level that meets the tracking requirements If the search window is selected, step 104 is performed.

[0080] In this embodiment, the tracking area can be set in the current image in advance. Since there are many methods for setting the tracking area in the prior art, it will not be repeated in this embodiment. It is assumed that the determined tracking area is a rectangular area (x s ,y s ,w s , h s ), where (x s ,y s ) is the coordinates of the upper left corner of the tracking area, (w s , h s...

Embodiment 2

[0136] The image tracking method in this embodiment is roughly the same as the image tracking method in Embodiment 1, the difference is that:

[0137] The first point: In order to reflect the edge characteristics and spatial position characteristics of the target, the target is divided into blocks according to the preset number of blocks in advance, and the block histogram of the target is calculated. Then the standard histogram of the target also includes the block histogram.

[0138] Wherein, the standard histogram of the target in the first embodiment is the global histogram of the target, and the histogram of the search window also refers to the global histogram of the search window. In this embodiment, the standard histogram of the target includes not only the global histogram of the target, but also the block histogram of the target. Assuming that the target is divided into M×N blocks, and the histogram includes the histogram and gradient direction histogram of each col...

Embodiment 3

[0194] The image tracking method in this embodiment may be consistent with the image tracking method in Embodiment 1, and may also be consistent with the image tracking method in Embodiment 2. The difference is:

[0195] In order to reduce the number of calculations of the histogram, improve the operation speed, to ensure the real-time performance of tracking, in this embodiment, the histogram (comprising block histogram and global histogram) described in step 104 in the embodiment one and embodiment two Figure) using Image 6 The method shown is calculated, Image 6 It is a flow chart of the histogram calculation method in Embodiment 3 of the present invention, and the process includes the following steps:

[0196] Step 601, pre-calculate the area integral histogram of the entire tracking area.

[0197] Wherein, when calculating the area integral histogram of the entire tracking area, the histograms of all areas in the entire tracking area with a preset corner of the entir...

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 an image tracking method comprising: performing confidence description for each search window, where a target may be appeared, in a whole tracking region by a classifier; if there is not a search window with a confidence satisfying request for tracking, calculating a histogram of each search window, matching the calculated histograms of the search windows with a standard histogram of the target, obtaining a matching result, and determining a tracking position of the target according to the matching result; if there is a search window with a confidence satisfying request for tracking, determining the tracking position of the target in the search window with a confidence satisfying request for tracking. In addition, the invention also discloses an image tracking system. The method and system provided in the invention further increases real-time performance of tracking by using a method in which a target classifier and the histograms are matched and combined.

Description

technical field [0001] The invention relates to image tracking technology, in particular to an image tracking method and system. Background technique [0002] In the current image tracking technology, the target tracking algorithm based on histogram matching is usually used to track the target, that is, to search for the best matching corresponding target in the area where the target may appear in a new frame of image, as the new position of the target. The specific process includes: determining each search window in the tracking area, matching the histogram of each search window with the standard histogram of the target, and taking the best matching search window as the new position of the target. Among them, considering that the size of the target in each frame of image may be different due to the distance movement of the target during the shooting process of each frame of image, that is, the scale of the target in each frame of image may be different, therefore, in In ad...

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 曾志王耀辉
Owner 中星智能系统技术有限公司
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