Real Time Face Recognition and Tracking System with Multi-Camera Arrangement

Abstract

Object tracking and face reorganization has received tremendous attention in the video processing community due to its various applications in video surveillance, traffic monitoring and so on. A single camera is not capable to scan 3d view of specified space. So, we use multiple cameras, placed in different sections of the area with overlapping region in field of view (FOV). Every camera will capture the video scene of itself FOV. The system is able to track human successfully by setting up correspondence between objects captured in multiple cameras. Thus, it saves the hectic job of manual tracking. There is a search window available for each object that gives the object’s trajectory. Tracking of object will be given by continuation of this process. For monitoring objects in areas like car parking, banks, hotels etc for security purpose, this system is best. Over the last couple of years, many algorithms and results have been presented for the problem of object tracking and recently the focus has been concentrated on real time person tracking with multiple cameras. Secondly, face detection is one of the best ways of identification. The main applications of automated face recognition are of biometric authentications and surveillances. Face recognition systems has became popular in biometric field as it is non intrusive and does not require the human interference. Up to that, there is no solution or technique that provides robust methods. This paper presents the detection of the face of the person, recognize and do tracking with use of multiple web cameras. Generally in daily camera security systems, cameras have been continuously remain on and large data storage is required in the system. In this real time object tracking system, Infrared sensors are used which indicates presence of person or object. Cameras will turn on only when object is detected by sensor after then the face recognition is carried out. It has capability of high speed processing and achieved low computational requirements. In similar areas, efficiency, accuracy, and speed of identification are the main tackled issues.