Sound Detection as an Aid to Increase Detectability of CCTV in Surveillance System

An operator in CCTV surveillance system is required to detect abnormal events over long working hours, and the events are intermittent, unpredictable and infrequent. Therefore, Operators often show lower performance than desirable. This paper proposes an automated surveillance system that integrates vision and audition to increase detectability. Sound surveillance system using TDOA (Time Difference of Arrival) can locate a sound source accurately, turn the camera towards it, and it has more advantages in reliability and cost-effectiveness than the existing surveillance system. The system is verified by conducting experiments in various environments.