Design Of Video Tracking System In Recording And Broadcasting Equipment

With the development of computer vision technology,intelligent equipment base on image processing is used widely.Especially in the embedded image analysis of video surveillance and other intelligent systems in artificial intelligence,multimedia education,security and other fields play an increasingly important role.Because of the increasing demand for video recording equipment,there is an urgent need for an intelligent classroom video recording and broadcasting system which does not require human intervention.It can make recordings without the operation of the director,these online courses make students feel in the real classroom.With the help of network servers and computers,synchronization courses and online courses can be achieved,thus forming a shared quality education resources system.This paper mainly studies the key techniques of detecting and tracking moving objects in intelligent recording and broadcasting system.Based on the characteristics of the actual classroom environment and the study of the existing target detection algorithm,this paper selected the background difference method.The Gaussian mixture model is used to model the background,and the updating mechanism of the background model is improved,which is more suitable for the detection of the ?big? and ?slow? moving target.Based on the background modeling,the foreground picture of the moving object is extracted by the background difference method.This paper focuses on the updating and optimization of the background modeling,the image binarization method and the moving object information extraction algorithm.Through simulation verification on VS2010 and embedded platform Hi3531 A,verify the feasibility of this algorithm.Combined with the actual situation of the classroom,this paper designs the method of detecting and locating the teachers in the intelligent recording and broadcasting system.In this paper,the system design based on Hi3531 A embedded platform is researched.By developing in Linux compiling environment,the detection and tracking algorithm is applied to the embedded platform,which realizes the video acquisition,video algorithm analysis,automatic tracking.