Research Of Target Detection And Shadow Detection Algorithm Based On Computer Vision

Computer vision is an interdisciplinary field which deals with how computers can be made for obtaining high-level understanding from digital images or videos.From the perspective of engineering,it seeks to automate tasks that the human visual system can do.Computer vision tasks include methods for acquiring,processing,analyzing and understanding digital images,and extraction of high-dimensional data from the real world in order to produce numerical or symbolic information,e.g.,in the forms of decisions.As a scientific discipline,computer vision is concerned with the theory behind artificial systems that extract information from images.The image data can take many forms,such as video sequences,views from multiple cameras,or multidimensional data from a medical scanner.As a technological discipline,computer vision seeks to apply its theories and models for the construction of computer vision systems.Intelligent video surveillance is a technology of using computer vision technology for video signal processing,analysis and understanding,without the need of human intervention,through the automatic analysis of the sequence of images to monitor and locate the changes in the scene.Intelligent Visual Surveillance System can distinguish and identify the objects from the numerous video images,and can analysis and extract useful information from the video images automatically,improving the performance of traditional intelligent video surveillance system.With the rapid advancement of the network technology and information technology,the intelligent monitoring technology in the field of pattern recognition has become one of the most hot academic topic in recent years with increasing attention.In the field of intelligent video surveillance,the detection and recognition of moving objects is a very important research topic.The traditional motion target detection technology includes optical flow method,frame difference method and background difference method,these algorithms can effectively detect the moving target.However,current techniques typically have one major disadvantage: shadows tend to be classified as part of the foreground.This happens because shadows share the same movement patterns and have a similar magnitude of intensity change as that of the foreground objects,their incorrect classification as foreground results in inaccurate detection and decreases tracking performance,therefore,how to get clean and precise object becomes an essential part.The main contents of this paper include the improvement of the mixture of Gaussian model and propose a shadow detection algorithm which combines the texture features with color features.Because the GMM algorithm is slow to converge,and the uniform updating rate used makes the process of building reference model timeconsuming and in some cases the background model cannot timely reflect the real background changes.This paper proposes an improved Gaussian mixture model which uses different learning rates At different time periods in the background update process,improving the rate of background modeling convergence and the detection rate of the target of interest.In this paper,a shadow detection algorithm based on fusion of color and texture features is proposed.The color-based feature is widely used in the process of cast shadow detection.However,when the moving target is similar to the background,the moving target is often classified as shadow,in order to overcome this defect,combining with texture features,this paper presents a cast shadow detection algorithm which is based on the fusion of color and texture features.Simulation experiments show the effectiveness of the algorithm.