Gradient operator in video after object detection by optical flow and morphological operation

Abstract

Image and video processing is a very important topic for research in the world. Its application pervade in the whole world, as perceived in video surveillance, anomaly detection, human action analysis etc. Nevertheless some drawbacks are there, as there is large storage required for storing the data and retrieval of a particular data and from the colossal storage it is very difficult or in some case it is even impractical by human resources. In such cases it is important to search for systems with high security and efficiency. This paper focuses on such a processing method. This paper describes how the edges of moving objects in the video will be found out. The first step is to detect the moving object, the result of which will contain some noises, which can be removed to a large extend by using morphological operation. The edges of the resultant output will give a clear picture of the moving object in the video. In this paper, apply optical flow method first to detect moving object, then morphological closing operation and then gradient edge detection techniques.