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

Multi-dimension background modeling method based on combination of textures and intensity characteristics

A technology of background modeling and feature fusion, applied in image data processing, instruments, closed-circuit television systems, etc., can solve problems such as light changes, large foreground objects, holes, etc., and achieve good robustness and accuracy

Active Publication Date: 2013-03-20
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the technical problems of light changes and large foreground object voids. To this end, the present invention provides a multi-scale background modeling method based on fusion of texture and intensity features for video surveillance

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
  • Multi-dimension background modeling method based on combination of textures and intensity characteristics
  • Multi-dimension background modeling method based on combination of textures and intensity characteristics
  • Multi-dimension background modeling method based on combination of textures and intensity characteristics

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0015] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in further detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0016] The multi-scale background modeling method based on texture and intensity feature fusion researched and developed by the present invention is a pattern recognition-oriented system implemented in C++ language using object-oriented design methods and software engineering specifications under the environment of microcomputer Windows XP.

[0017] figure 1 It is the implementation flowchart of the method of the present invention. Below to figure 1 An example is used to illustrate some specific implementation processes in the present invention. The method of the present invention utilizes a multi-scale background modeling method based on texture and intensity feature fusion, and its specific steps are:

[0018] 1. Per...

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 multi-dimension background modeling method based on combination of textures and intensity characteristics. The method comprises the following steps: performing multi-dimension conversion on the input image sequences; representing the image in each dimension by using the same-dimension centrally symmetrical partial ternary mode operator; calculating the probability that the mode in each dimension image belongs to the background by using the self-adaptive mode kernel density method; regarding the probability density distribution with the accumulated weight exceeding the texture proportional threshold as the background, and calculating the probability that the mode belongs to the background; representing the intensity image of each dimension as the intensity image; regarding the Gaussian model with the accumulated weight exceeding the intensity proportional threshold as the background; calculating the probability that the intensity value belongs to the background by using the simplified mixed Gaussian model; mixing the image textures and image intensities with the same dimension; mixing the images with different dimensions so as to obtain the final probability belonging to the background; when the probability is greater than the background division threshold, belonging to the background; and on the contrary, belonging to the foreground.

Description

technical field [0001] The invention belongs to the technical field of intelligent video monitoring, and in particular relates to a multi-scale background modeling method based on fusion of texture and intensity features in video monitoring. Background technique [0002] Background modeling plays an important role in video surveillance. Background modeling can be used to extract foreground moving objects, and foreground object extraction is the first step in moving object tracking and behavior analysis. Its accuracy directly affects subsequent Work. In practical applications, a very challenging problem in background modeling is the processing of dynamic backgrounds, including: rippling water surfaces, swaying trees, and camera shakes. Another challenging problem is the variation of lighting. [0003] In recent years, a large number of background modeling methods have been produced to solve the above problems, but the effect in complex backgrounds is not satisfactory. For ...

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): G06T7/00H04N7/18
Inventor 王春恒张重肖柏华周文
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
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