Title: An intelligent video surveillance system for rapid detection of abandoned object in both indoor and outdoor environments

Authors: R. Ganapathyraja; S.P. Balamurugan

Addresses: Department of Computer and Information Science, Faculty of Science, Annamalai University, Chidambaram, Tamil Nadu, India ' Department of Computer and Information Science, Faculty of Science, Annamalai University, Chidambaram, Tamil Nadu, India

Abstract: Video surveillance is an essential tool of the security system for monitoring both public and private places for security measures against crime events such as terrorist attacks and bomb explosions. In recent years, the number of terrorist attacks has increased across the world as terrorists hide explosives in luggage and leave them in public places. The conventional video surveillance in the above places does not prevent terrorist attacks. Nevertheless, this existing surveillance system only records events. Manual analysis of the video footage takes place only after an unwanted incident happens. It is not helpful to prevent untoward incidents. Therefore, an efficient automatic detection system is needed to prevent hidden explosive attacks (abandoned objects) and suspicious loitering human activities. This article proposes two algorithms for detecting abandoned objects and suspicious loitering humans in real-time scenarios in the AOD system. In this system, addressing the issues has different stages that include moving foreground object segmentation, detecting the stationary object, and Identifying the owner.

Keywords: video surveillance; spatial-temporal; contour; stable region; stationary object; suspicious object; abandoned object.

DOI: 10.1504/IJCVR.2026.153139

International Journal of Computational Vision and Robotics, 2026 Vol.16 No.3, pp.330 - 346

Received: 12 Dec 2023
Accepted: 16 Jan 2024
Published online: 23 Apr 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article