Research On Intelligent Monitoring Technology Of Water Surface Floats Based On Background Subtraction

Intelligent video monitoring is an automatic video monitoring technology,which includes the detection,analysis and interpretation of the corresponding behavior of moving targets.The detection of targets in video streams is the essential technique in intelligent video monitoring.Background subtraction algorithm is the most widely used method in target detection.In recent years,with the acceleration of industrialization,a large number of water surface floats appear on the surface of lakes,rivers and reservoirs.In order to solve this problem and save the cost of human and material resources as much as possible,the intelligent monitoring of water surface floats has become an urgent problem to be solved.In this paper,we analysis the background subtraction algorithm in depth and propose some solutions for several typical difficulties in the detection and segmentation of floating objects.The main work of this paper is as follows:(1)Based on the classical Gaussian mixture model(GMM),this paper proposes an improved GMM-based automatic segmentation method(IGASM)for the detection of water surface floats,where the background updating strategy is improved to segment the water surface floats more effectively.Besides,the GMM results are mapped into an HSV color space,and a light-shadow discriminant function is applied to solve the problems of light and shadow.Then,a morphological method is used to smooth the extracted foregrounds.(2)The classical GMM method probably detects incorrectly and only classifies the pixels on the time scale,and thus it cannot monitor the water surface floats effectively.To solve this problem,Graph Cuts algorithm is applied in this paper to optimized the segmentation results according to the spatial information of video images.(3)In order to solve false detection caused by camera jitter in the process of monitoring,an image alignment method is proposed in this paper.This method firstly finds feature points with SURF algorithm,then an improved RANSAC algorithm is used to calculate homography matrix,and the corresponding relationship between the two images is established to compensate the detection error caused by the above reasons.