A method of target detection image processing

A technology of image processing and target detection, which is applied in the field of image processing, can solve problems such as waste of computing time and storage space, discrimination confusion, etc., and achieve the effect of saving storage space and reducing computing time

Inactive Publication Date: 2016-07-06
NANJING UNIV OF SCI & TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Such an algorithm has the following defects: (1) Repeated calculations on non-interest areas (areas without target points) waste calculation time and storage space; (2) When the background light changes suddenly, the background points If there is confusion with the discrimination of foreground points, the background detection will become the foreground target; (3) When the detected target enters the field of view, if the target is still, it will automatically integrate the target into the background and become a part of the background

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 method of target detection image processing
  • A method of target detection image processing
  • A method of target detection image processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] The basic principle of the mixed Gaussian background update method of the present invention is based on the expectation maximization algorithm (ie EM algorithm), estimates the background model of the image with the method of maximum likelihood estimation, and then compares the background model with the image to obtain the foreground target . Since the brightness of each pixel obeys a Gaussian distribution, such as the brightness value I(x,y)~N(μ,σ) of a point (x,y), the attributes of each pixel can be obtained from the Gaussian distribution that the point obeys. μ and σ, so the Gaussian distribution is used to update and match when the mixed Gaussian background update algorithm builds the background model and matches the target points.

[0019] For the pixel value J of the Mth pixel of the T frame image Μ,Τ The probability distribution of

[0020] P ( J M , ...

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 target detection image processing method. An image is divided into an interest region and a non-interest region for processing; the image is processed by different Gaussian distribution numbers and weight update rates in different regions; meanwhile, the situations that the background generates mutation and the foreground target is quiet can be processed by updating different background update rates; the capability on detecting an infrared target by a hybrid Gaussian background update algorithm can be effectively improved; and a good effect is obtained by a simulation experiment.

Description

technical field [0001] The invention belongs to image processing technology, in particular to an image processing method for target detection based on an improved mixed Gaussian background update algorithm. Background technique [0002] Nowadays, infrared target detection is a hotspot and difficulty in image processing. How to detect infrared targets with limited information in complex backgrounds, and the background is not completely unchanged, is the difficulty of studying infrared target detection. For the special characteristics of infrared target detection, the mixed Gaussian background update algorithm is obviously superior to target detection methods such as optical flow method or background frame difference method (1. Li Mingzhi, Ma Zhiqiang, Shan Yong, Zhang Xiaoyan. Adaptive Gaussian mixture model under complex conditions Background Update. Computer Applications, 2011, 31(7):1831-1834). This is because the background frame difference method cannot deal with the ch...

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 & AuthorityPatents(China)
IPC IPC(8): G06K9/46G06T7/00
Inventor顾国华王厚周霞胡楷钱惟贤陈钱徐富元任侃隋修宝何伟基路东明于雪莲李宏哲
OwnerNANJING UNIV OF SCI & TECH