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

Moving object detection method

A technology of moving objects and detection methods, which is applied in the field of video or image sequence processing, and can solve problems such as high noise and poor robustness.

Active Publication Date: 2013-01-30
SUZHOU UNIV
View PDF2 Cites 49 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The purpose of the present invention is to provide a moving target detection method to solve the problems of excessive noise and poor robustness in traditional background modeling, improve the accuracy and speed of moving target detection, and make it meet the requirements of intelligence

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
  • Moving object detection method
  • Moving object detection method
  • Moving object detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] Embodiment 1: a kind of moving object detection method, comprises background modeling and foreground detection, described background modeling sees figure 2 ,include:

[0058] (1) Model building process: obtain the video of the scene to be detected, and obtain t frames of images according to the video content, where t is an integer greater than or equal to 1;

[0059] The color value of each pixel in the image is regarded as a statistical random process:

[0060]

[0061] Among them, I(x, y, i) represents the color value of the pixel point (x, y) in the i-th frame of the image, that is, at the time i. The resulting type II fuzzy mixture Gaussian model:

[0062]

[0063] Among them, ω j,t is the weight of the jth Gaussian component at time t, is the probability density function of the jth Gaussian component at time t, P(X t ) represents the probability of occurrence of the observed pixel value X at time t;

[0064] (2) Model update process: If the currently...

Embodiment 2

[0074] Example 2: Mixed Gaussian background modeling with Type II blurring. In this example, aiming at the problem that the dynamic scene has more noise caused by the statistical defect of the pixel distribution of the ordinary mixed Gaussian model itself, the technical means of fusion type 2 fuzzy is used to describe and model the pixels of the dynamic scene more accurately, which is extremely Greatly removes foreground noise.

[0075] (1) First, the weakness of the mixed Gaussian model is analyzed through experiments: a certain background pixel sequence of the video sequence does not strictly satisfy the mixed Gaussian distribution, such as Figure 5 shown.

[0076] Figure 5 The odd-numbered columns represent the scene of the experimental data set, and the even-numbered columns represent the distribution law of a certain pixel in the scene. Obviously, none of the pixels strictly satisfy the Gaussian distribution, especially (d), (h), (j) , (p) and other occasions where t...

Embodiment 3

[0082] Embodiment 3: Mixed Gaussian background modeling based on block-level features. Aiming at the traditional block-level feature extraction speed is slow, the extracted foreground is not smooth enough, and the foreground loophole problem in the first embodiment, a novel feature extraction method based on discrete cosine transform is used to construct a pseudo image. The figure is modeled with a Gaussian background, which has a very complete foreground detection effect and has a certain inhibitory effect on noise.

[0083] (1) Figure 8 It is the result of the mixed Gaussian background modeling method of fusion type II blurring in embodiment 2, and it is found that it has defects, such as Figure 8 As shown, the result has a large loophole:

[0084] Therefore, it is necessary to repair the existing loopholes. Through experiments, it is found that the integrity of foreground objects can be guaranteed by using block-level features. Common methods have defects such as insuf...

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 moving target detection method, which comprises the steps of background modeling and foreground detection, wherein the background modeling is performed by using a gaussian mixture model, and an operation of model updating is performed so as to obtain B gaussian distributions, for a pixel in a frame image to be detected, if in the B gaussian distributions subjected to sequencing, at least a gaussian distribution is matched with the current pixel value, the pixel is a background pixel, otherwise, and the pixel is determined as a foreground pixel. According to the invention, foreground noises can be effectively filtered so as to obtain an extremely clean background, thereby, an efficient denoising effect is kept, and the accuracy and completeness of foreground object detection are enhanced.

Description

technical field [0001] The invention relates to the processing of video or image sequences, in particular to a method for detecting moving objects in the video or image sequences. Background technique [0002] With the development of the economy and the improvement of people's living standards, the requirements for property safety and personal safety are also increasing. With the continuous improvement of social security awareness, more and more video surveillance systems are used in places such as campuses, communities, roads, and densely populated areas. At present, most of the monitoring systems adopt the relatively primitive manual on-duty method, which has two disadvantages: 1. Due to the interference of people's quality, physiological needs and other factors, it is difficult to ensure that some monitoring systems will not be missed during manual on-duty monitoring. Key information, which leads to safety problems; 2. In most cases, it is only necessary to pay attention...

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
Inventor 龚声蓉张居涛刘纯平季怡王朝晖潘林林
Owner SUZHOU 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