Adaptive firearm recognition algorithm in a dynamic background
Noel Christian A. Corales, Matt David B. Cruz, Ivan Mikael F. David, Alvin Garcia, Michael Benedict C. Isidro, Christian John D. C. Ng, Kanny Krizzy D. Serrano · 2017
The modern-day video-surveillance system is a significant advancement in technology used in various applications with the aim of automatizing the recognition of dangerous situations and helping human security officers to make the most appropriate decisions in order to ensure the safety of the general population. With the ever-growing crime rate in the Philippines, the government made counter measures to ensure the safety of the citizens. CCTV (closed-circuit television) cameras were installed in public places in which the government aims it to help deter crime and solve investigations. However, index crimes is still growing even though the CCTV cameras were installed in numerous places since current surveillance and control systems still require human supervision and intervention. This study proposes to empower the CCTV cameras with the use of an automatic firearm detection system in videos appropriate for surveillance purposes. With the adaptation of existing object detection techniques, an algorithm will be designed according to the given specifications of the CCTV cameras installed in the Philippines. The proposed algorithm will trigger an alarm signal when a firearm was to be detected. This study aims to help reduce the number of crimes that involves firearms and to ensure a safer community for the people with the use of the proposed algorithm.