Fuzzy logic based human detection for CCTV recording application

Different categories of crime, location, victim and offender details have a national security concerns. Premises surveillance needs are growing. One of the surveillance systems is the closed-circuit television (CCTV) that able to provide information about the crime. With the increase in wide usage of the system, more technical flexibility for CCTV systems is required. Human detection techniques can play big role in this field and provide more efficient working CCTV system. The system work based human detection resulting in most efficient in saving recording space. Part of the human detection is a video motion detection system that works to trigger the system to increase its frame rate recording to a predetermined value. The system also can be useful in triggering an alarm only in moments of a human appearance. The paper presents a design of a human detection system utilizing CCTV cameras based on fuzzy logic model. The propose system aiming to has a capability of identifying certain objects in line of sight and determine if the object is a human or not. The propose system is a real time system for human detection, tracking, and verification in such challenging environments. The system integrates several computer vision techniques in order to design an algorithm that is capable of processing the digital images to perform its functions that include human detection, object tracking, and motion analysis. The testing results provide very encouraging outcome for the proposed system