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

Method for monitoring moving object in video image

A technology of moving objects and video images, applied in the field of visual analysis, can solve the problems of reducing the quality of background images, misjudgment, and not using prior knowledge of regional continuity

Active Publication Date: 2018-02-09
HENAN POLYTECHNIC UNIV
View PDF4 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But this method also has the following two problems: (1) use l 1 The norm realizes the constraint of sparsity. This method treats each pixel independently, without using the prior knowledge of the regional continuity of the foreground moving target, and will misjudge some scattered non-target large noise as the foreground moving target.
According to the physical meaning of the singular value, the large singular value contains the main information of the image, which will inevitably reduce the quality of the decomposed background image

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
  • Method for monitoring moving object in video image
  • Method for monitoring moving object in video image
  • Method for monitoring moving object in video image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0071] Such as figure 2 As shown, the present invention provides a kind of method for monitoring moving target in video image, and this method comprises the following steps:

[0072] 101. Construct an observation matrix D of the video image to be processed;

[0073] Specifically, in this step 101, grayscale and vectorization operations may be performed on each frame of the acquired surveillance video sequence to construct an observation matrix.

[0074] For example, the obtained N frames of surveillance video images are grayscaled, and the N frames of images obtained after the grayscale are denoted as I 1 ,...,I N , the resolution of each frame image is recorded as a×b, that is, I i ∈ R a×b ,i=1,...,N,R a×b Represents the space of real numbers of size a×b.

[0075] and the I 1 ,...,I N Sequential vectorization to construct the observation matrix D∈R M×N , where M=ab,

[0076] R M×N Represents a real number space whose size is M×N, and the specific operation is as f...

Embodiment 2

[0128] In this embodiment, the method of the present invention is used to detect and track moving objects on the monitoring video of the CDNET2014Office scene.

[0129] 201, the obtained 100 frames of three-channel color surveillance video images are grayscaled, and the 100 frames of images obtained after the grayscale are denoted as I1 ,...,I 100 , the resolution of each frame image is recorded as 360×240;

[0130] 202, Will I 1 ,...,I 100 Sequential vectorization to construct the observation matrix D∈R 86400×100 , where 86400=360×240, the specific operation is as follows:

[0131] D=[Vec(I 1 ),…,Vec(I 100 )]∈R 86400×100

[0132] Vec(I i ) represents a vectorized function that takes the matrix I i The columns of are sequentially concatenated from left to right into an 86400×1 vector.

[0133] 203. Based on the theory of robust principal component analysis, a cost function is established:

[0134]

[0135] in:

[0136] L∈R 86400×100 Represents the low-rank matr...

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 method for monitoring a moving object in a video image, and the method can be used for an occasion where a monitored video is intelligently analyzed. The video image is formed through superposing a background image and a foreground moving object. According to high correlation among background image sequences and a sparsity characteristic existing in a foreground image, arobust principal component analysis method can be used to realize separation of a background and a foreground. A weighted nuclear norm is used as a low rank constraint of a matrix so that sizes of a compression threshold and a corresponding singular value form a monotone decreasing relationship, and a large singular value can be compressed in a small amplitude mode. A structure sparse norm is usedas a foreground sparse constraint and priori knowledge of space area continuity of the foreground moving object is effectively used. The weighted nuclear norm and the structure sparse norm form a newcost function and an alternating direction multiplier method is used to carry out optimization so that the moving object in the monitored video is effectively and accurately detected and tracked.

Description

technical field [0001] The invention relates to visual analysis technology, in particular to a method for monitoring moving targets in video images. Background technique [0002] Moving object detection is to separate the background image and the moving object image (also known as the foreground) in the surveillance video frame sequence, which is an important basic step in intelligent video analysis. Accurate detection of moving objects is of great significance to the completion of subsequent high-level computer vision tasks (such as behavior recognition, scene analysis, traffic control, etc.). In recent years, scholars at home and abroad have carried out extensive research on this subject and proposed many algorithms. However, moving target detection is still a research difficulty and hotspot in the field of machine vision because it faces many challenges such as illumination changes, dynamic background, camera shake, and algorithm real-time performance. [0003] Backgrou...

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
CPCG06T7/20G06T2207/10016
Inventor 张延良李兴旺李赓卢冰
Owner HENAN POLYTECHNIC 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