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

A video moving object detection method based on non-local self-similarity

A technology of self-similarity and moving objects, applied in image analysis, image enhancement, instruments, etc., can solve problems such as false detection, incomplete moving objects, hollowing out, etc., to improve false detection problems, improve accuracy and efficiency, short time effect

Inactive Publication Date: 2019-06-07
TIANJIN UNIV
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Comprehensive analysis of several problems existing in existing video moving object detection algorithms: Misdetection of moving backgrounds (such as shaking leaves, ripples on the water surface, etc.) as foreground; detected moving objects are incomplete and hollow

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 video moving object detection method based on non-local self-similarity
  • A video moving object detection method based on non-local self-similarity
  • A video moving object detection method based on non-local self-similarity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The technical problem to be solved by the present invention is to realize video moving object detection by adding non-local self-similarity constraint items on the basis of matrix representation.

[0036] Batch video moving object detection algorithms based on matrix decomposition usually include foreground sparse items, smooth items, background reconstruction representations and background low-rank constrained items. Usually there are the following constraints to optimize the objective function:

[0037]

[0038] Or transformed into an unconstrained optimization problem:

[0039]

[0040] In formula (2), F is the foreground matrix and F ij ∈{0,1}, a value of 1 indicates that the point is a foreground point, otherwise it is a background point; O is the vectorized matrix of the input video sequence; B is the background matrix and B ij ∈[0,1]. is the background reconstruction representation item, with F ij The value is opposite, this item means that when F i...

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 relates to a video moving object detection method based on nonlocal self-similarity. According to the method, improvement is made on a batch video moving object detection algorithm DECOLOR. The method comprises the following steps: segmenting each frame of image in a video sequence into image blocks; calculating the nonlocal self-similarity value of each image block; getting a nonlocal self-similarity matrix S; using a vectored nonlocal self-similarity matrix Q to constrain a foreground matrix F to get nonlocal self-similarity constraints; adding the nonlocal self-similarity constraints to an objective function of DECOLOR to get a new objective function; solving an unconstrained minimization problem for the new objective function; vectoring each frame of image of a video sequence to be processed into an input matrix O; and performing iterative calculation to get a new foreground matrix F. The method has the advantages of high calculation speed and good effect.

Description

technical field [0001] The invention relates to an efficient and fast video moving object detection method in the fields of computer vision, pattern recognition and the like, in particular to a video moving object detection method using non-local self-similarity. Background technique [0002] Video automatic analysis technology plays an important role in video surveillance, augmented reality, car navigation, etc. Automatic video analysis technology [1] includes three steps: moving target detection, moving target tracking and moving target behavior analysis. As the basis of automatic video analysis technology, moving object detection refers to extracting moving objects from video in real time to provide prior conditions for subsequent processing (such as: object classification, etc.). Therefore, how to detect moving objects in real time and accurately is very important. However, in the existing video moving object detection technology, the shaking of objects in the video ba...

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): G06T7/246
CPCG06T2207/10016
Inventor 叶丽庞彦伟
Owner TIANJIN 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