Object Tracking and Suspicious Activity Identification during Occlusion

Abstract

Rising criminal activities and demand of robust security solutions, detection and tracking of every minute detail of suspicious activity or object has become a topic of interest for researchers all around the world. In this paper, we propose an approach based on Digital Image and Video processing to detect and track the motion of multiple objects during the phenomenon of occlusion and activate an alert if an object is dropped for a long period of time in the region of concentration of camera. The proposed method can be utilized in video surveillance system and the method has been verified through extensive experimentation for multiple video.

Detecting possibly threatening conspicuous human behaviour in airport terminals via movement categorization

“…This work focuses on the movement behaviour. Yakkali et al (2018) consider the movement of objects in video surveillance data. Suspicious movement is assumed to occur if a detected object does not move for a number of frames w.r.t.…”