Unlock instant, AI-driven research and patent intelligence for your innovation.

A Target Tracking Method Based on Contour Gradient

A target tracking and gradient technology, which is applied in the field of target tracking based on contour gradient, can solve the problems that feature matching calculation is difficult to achieve real-time, the reference image is random, and the response speed of target tracking is affected.

Active Publication Date: 2020-09-25
CHINA FORESTRY STAR BEIJING TECH INFORMATION CO LTD
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0012] (2) The feature points of the obtained reference image are arbitrarily large, usually contain less image information, cannot represent image features well, and the reliability and stability are not high, so the tracking algorithm does not have good robustness
[0013] (3) It is difficult to achieve real-time feature matching calculation, which affects the response speed of target 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
  • A Target Tracking Method Based on Contour Gradient
  • A Target Tracking Method Based on Contour Gradient
  • A Target Tracking Method Based on Contour Gradient

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0102] Embodiment 2: The calculation of the similarity in Embodiment 1 involves the convolution operation of the gradient eigenvalues ​​of the feature template of the reference image and the feature matrix of the target image within a certain window range. Although based on the definition that the gradient value of the feature gradient corresponds to (0,0) at non-edge points, this calculation process has greatly reduced the amount of calculation compared with the traditional feature matching method. However, the calculation of convolution is mechanical and complicated, and still consumes a lot of time. As a result, it is difficult to achieve real-time calculation of the above-mentioned matching similarity.

[0103] Therefore, here is a fast calculation method, using fast Fourier transform (FFT) for convolution operation, and converting the similarity to the frequency domain. The specific method is as follows:

[0104] Step1. For any multi-scale and multi-angle feature template...

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 present invention specifically discloses a target tracking method based on contour gradient, by segmenting the candidate target from the initial frame of the input video or image sequence, and extracting the reference image; extracting the reference image contour as a standard feature template, and performing scale and angle Transform to obtain a multi-scale and multi-angle template sequence; input the next frame of the video or image sequence, and extract the gradient feature of the target image; perform a sliding window with a step size of 1 on the multi-scale and multi-angle template sequence on the feature image of the target image Scan and calculate the similarity between the two matches; according to the matching position, scale factor, angle factor and other information, segment the sub-region where the target is located in the target image as the reference image for the next detection and update it; until the video or image sequence is all Frame detection is complete. The purpose of the present invention is to provide a method for tracking a target based on contour gradients to adapt to scale changes and / or angle changes of the target and improve the accuracy of target tracking.

Description

technical field [0001] The invention belongs to the technical field of computer image processing, and specifically discloses an object tracking method based on contour gradients. Background technique [0002] In security monitoring tasks, in order to achieve the requirement of automatic target tracking, it is often necessary to determine the target position of the region of interest in the video or continuous image sequence, and correspond to the target position in each frame of image. [0003] The research content of target tracking technology is mainly divided into the following two aspects: 1. To detect, track, identify and extract the required information for the moving target in the captured video sequence, such as the trajectory of the target and related motion parameters such as speed, acceleration, position, etc. Second, use the acquired motion parameters to predict and estimate the target to assist in decision-making. Therefore, the precise extraction of the featu...

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): G06K9/00
CPCG06V20/46
Inventor 李波左春婷蔡宇黄艳金
Owner CHINA FORESTRY STAR BEIJING TECH INFORMATION CO LTD