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

A fast and robust object tracking method based on k-sparseness

A target tracking and sparse technology, applied in the field of fast and robust target tracking based on K-sparse, can solve the problems of not meeting the real-time requirements of target tracking and large amount of calculation

Active Publication Date: 2021-07-27
SHAOGUAN COLLEGE
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, since the sparse cooperative robust target tracking method uses LASSO to solve l 1 Optimization problem, and the LASSO solution process has a large amount of calculation, which cannot meet the real-time requirements 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 fast and robust object tracking method based on k-sparseness
  • A fast and robust object tracking method based on k-sparseness

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] see figure 1, the present invention provides a fast and robust target tracking method based on K-sparse, comprising the following steps:

[0073] S1: Obtain the first frame image and the current frame image, and obtain the first frame target area of ​​the first frame image.

[0074] In the step S1, the first frame image and the current frame image are input into the relevant processing device through an external camera device or storage device, so that the processing device can acquire the continuous first frame image and the current frame image; and, the acquired The target area of ​​the first frame essentially obtains the position parameter of the area occupied by the target to be tracked in the first frame image, and the position parameter is obtained by inputting to the processor.

[0075] S2: Establish a positive and negative template dictionary A according to the first frame image and the first frame target area, and generate a redundant dictionary D according to...

Embodiment 2

[0131] The fast robust target tracking method based on K-sparse in this embodiment is basically the same as the fast robust target tracking method based on K-sparse in Embodiment 1, the only difference is that this embodiment adds Steps S10 and S12, in order to update the positive and negative template dictionary and redundant dictionary as the basis of discrimination according to certain rules in the process of realizing the target tracking, so as to ensure the applicability and accuracy of the basis of discrimination, which is more conducive to the target in subsequent images Tracking is analyzed and processed, and more accurate tracking results can be obtained. see figure 2 Compared with the target tracking method in embodiment 1, the K-sparse fast and robust target tracking method in this embodiment 2 further includes the following steps:

[0132] S10: Determine whether the frame number of the current frame image is an integer multiple of N, if yes, execute step S11, oth...

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 relates to a fast and robust target tracking method based on K-sparseness. The sparse matrix is ​​obtained by using the positive and negative template dictionary, and the positive and negative template dictionary is obtained by using the sparse matrix, that is, K-sparse 1 The optimization problem realizes that when the sparsity is relatively small and the signal-to-noise ratio is the same, the speed of solving sparse coefficients is more than ten times that of the LASSO algorithm; and, by dividing the current frame image into multiple image samples, and using regularization Based on the final positive template matrix and negative template matrix, the reconstruction error is calculated, and the target tracking in the current frame image is realized by using the reconstruction discriminant error parameters and the samples obtained by comparing the histograms, which also improves the target tracking accuracy. Accuracy, can get more accurate tracking results, and improve the real-time performance of tracking.

Description

technical field [0001] The invention belongs to the technical field of video image processing, in particular to a fast and robust target tracking method based on K-sparse. Background technique [0002] Target tracking is a method of estimating the target state in a video or image sequence, which plays a very important role in many machine vision applications, such as motion analysis, behavior detection, video surveillance, traffic monitoring, medical image processing, etc. effect. Although the target tracking technology has made great progress in recent decades, how to provide a robust and fast target for some complex dynamic scenes due to brightness changes, camera shake, target occlusion, and target shape position changes? Tracking methods remain a huge challenge in the field of computer vision. [0003] Currently, for target visual tracking, there are mainly two tracking models—morphological model and motion model. Among them, the morphological model is used to represe...

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): G06T7/246G06K9/46
CPCG06T7/246G06V10/40G06V10/513
Inventor 杨森泉周永明陈景华陈锦儒文昊翔罗欢
Owner SHAOGUAN COLLEGE