Research On The Foreground Detection Algorithm Of Suppressing Light Mutation And Shadow In Intelligent Monitoring System

Foreground detection is one of the important steps for the intelligent monitoring system to obtain external information,and it has a wide range of applications in the fields of security systems,smart homes,intelligent transportation,and national defense security.With the continuous innovation of intelligent monitoring systems,foreground detection technology has gradually become a research focus in recent years.As the basic technology of intelligent monitoring system,foreground detection has an important position and research value.It is the prerequisite for high-level processing of subsequent video sequences such as target tracking and behavior recognition.At present,foreground detection problems in complex scenes,such as sudden changes in local illumination,shadow interference,etc.,are research hotspots at home and abroad.In this paper,the foreground detection and shadow suppression under the sudden change of local illumination are studied.main tasks as follows:Aiming at the problem that the existing foreground detection algorithms are more sensitive to sudden light changes in the scene,a fusion and improved LBP operator mixture Gaussian model algorithm is proposed.Based on the LBP operator,this operator model combines spatial and temporal information,adds central pixel information,and introduces an adaptive illumination factor,so that the threshold can be adaptively changed with different illumination.At the same time,the Gaussian model algorithm and the LBP algorithm complement their respective shortcomings in foreground detection,so that the algorithm in this paper solves the "hole" problem in the Gaussian model detection while retaining the accuracy of the LBP algorithm to detect edge information.This paper verifies the effectiveness of the algorithm by conducting experiments on common experimental scenes and special scenes with sudden changes in local illumination.Aiming at the problem of shadow interference in foreground detection,a gradient algorithm based on HSI color space and brightness ratio is proposed.In this paper,the brightness ratio gradient is combined with the color characteristics of the H and S channels of the shadow HSI space to remove the shadow,so as to achieve the detection purpose and ensure the integrity of the foreground.Experimental results show that the shadow suppression algorithm proposed in this paper achieves the detection effect,removes moving shadows well,while retaining the foreground target.Based on the above work,using the classic MATLAB software design platform interface,and the Open Cv2.4.9 open source function library to provide interface functions for image processing,design and implement the human-computer interaction interface of the video detection processing module system in the intelligent surveillance video system,which is convenient Users have better control over the foreground detection and subsequent target tracking process.