Edge Detection Based Motion Tracking

Abstract

Tracking and Detection of objects in group of video is very useful in computer vision. This has numerous application in human computer interaction, robotics, surveillance systems and other fields. All of these systems necessitate real-time processing, and finding a way that is both efficient and simple. This work presents robust and fast approach to identify and to track the moving objects. The majority of present methodology is centered on tracking the edge detection of mobility via the fix edges. Image capturing, background subtraction, and Canny edge detection can all be used to detect moving objects. The background subtraction technique, as used in my techniques, is based on directly subtracting two consecutive frames to derive the difference image. The difference image denotes the locations where a moving item was in frame N and where the object is now in frame N+1. The results reveal that, in addition to its efficiency, the suggested method is capable of overcoming problems such as variations in brightness and changes in background over a time.