Optical Flow Based Moving Object Detection and Tracking for Traffic Surveillance

automated motion detection and tracking is achallenging task in traffic surveillance. In this paper, a systemis developed to gather useful information from stationarycameras for detecting moving objects in digital videos. Themoving detection and tracking system is developed based onoptical flow estimation together with application andcombination of various relevant computer vision and imageprocessing techniques to enhance the process. To removenoises, median filter is used and the unwanted objects areremoved by applying thresholding algorithms inmorphological operations. Also the object type restrictions areset using blob analysis. The results show that the proposedsystem successfully detects and tracks moving objects in urbanvideos.

Keywords:

optical flow estimation , moving object detection , tracking , morphological operation , blob analysis