基于高斯背景模型的红外人体运动目标检测技术

针对红外监控中人体运动目标的空洞和拖尾问题,提出了一种基于高斯模型的运动目标检测方法。首先,介绍了红外图像的预处理;其次通过与其他经典的人体运动目标检测算法比较与综合,引入高斯模型,建立背景图像的自适应模型。该种模型主要使用了拟合修正的方法处理了红外监控背景图像中的差分信息,过滤图像中的噪声等相关外部环境干扰因素,从而更新红外图像中的背景信息,提高了红外监控系统图像中人体运动目标的检测清晰度,并进一步提高了红外监控图像的精度。同时,还对该方法进行了必要的仿真实验。仿真结果表明,提出的方法可以准确地检测红外监控图像中的人体运动目标,较好地避免了人体运动速度过快或过慢所产生的拖尾或空洞现象。

Infrared human motion target detection based on gauss background model

Aiming at the holes and tails of human moving targets in infrared surveillance,a moving target detection method based on Gauss model is proposed in this paper.Firstly,this the preprocessing of infrared image is introduced.Secondly,by comparing and synthesizing with other classical human moving object detection algorithms,the Gauss model is used to build the adaptive model of background image.Fitting correction is used to deal with the differential information in the background image of infrared surveillance system,and filter the noise in the image and other related external environmental interference factors,so as to update the background information in infrared image,improve the detection clarity of human moving object in infrared surveillance system image,and further improve the detection clarity of human moving object in infrared surveillance system image.The accuracy of infrared surveillance image is achieved.At the same time,the necessary simulation experiments are carried out.The simulation results show that the proposed method can accurately detect human moving targets in infrared surveillance images and avoid the tailing phenomenon caused by the change of human moving targets’state.