| This task study of automatically tracking for moving objects in the video monitoring system,design of low-cost resources automatic monitoring system based on study of the algorithm about moving object detection and tracking to achieve unmanned operation to control purposes.Use ARM S3C2410 as the controlling microprocessor core,running Linux operating system to building development platform,to collection video signal, data transfer,and image processing.Based on study of the algorithm about moving object detection and tracking, analysis of motion detection and testing procedures improve the background difference to detect moving objects ,adopt integration frame difference thinking background difference law to on exercise objects detection, choose value 50 to binarization processing the results show that eliminate some noise, improve the detection accuracy. Extract the centroid of moving region as the reference position of moving objects,use Kalman filter to estimate the prediction of next state of motion, convenience of calculation, to achieve better results.Taking into account the actual situation, when the cameras turn to track moving objects, it stop collecting images, when the camera rotation certain angle, suspension of rotation for capture continuous video images for analysis.ARM S3C2410 as the controlling microprocessor core, design microprocessor hardware system, SDRAM,Flash memory circuit,wireless communication module circuit.Design vivi and transplanted the cut-off of Linux operating system to development platform.Based on the Video4Linux core to collection video signal, the video signal processing,design USB camera driver,to achieve moving object detection and tracking algorithm programming. This task study of simple and feasible algorithm for moving object detection tracking, designed hardware and software platform.In a simple scene,achieve the more accurate detection and tracking for moving objects. |
PDF Full Text Request