Efficient method for detecting and tracking moving objects in video

摘要

Detection and tracking of moving object in the video has turned into a fascinating zone of exploration in the field of computer vision and has wide applications in fields like video surveillance, service robots, public security and target recognition. Although researches have proposed many approaches for object detection, robustness remains a huge challenge. In this paper, we proposed hybrid method of object detection using motion estimation and tracking by parallel kalman filter. Detection of the moving object is performed by analyzing the moving parts by corner and the shape feature extraction. The parallel kalman filter is used to tracks the object that is detected by calculating the motion estimation in video. The object tracking using the shape and the corner feature helps in providing the extract tracking of the object.