Fuzzy analysis and adaptive anthropometry model for object identification in surveillance system

Abstract

The proof of identity in many cases like road traffic accidents, bomb blasts and other terrorists‟ attacks are most frequently achieved by comparing the individual‟s appearance to a previously captured image. This process is however cumbersome thus necessitating us to propose a system that automatically detects and focuses on the human face and inputs it to the face recognition system for further processing. The methodology uses a combination of temporal differencing and background subtraction method for object detection. It is then followed by a template-matching classification algorithm which uses the object silhouette followed by the contour tracing algorithm. After object detection and classification, we compute the anthropometric  measurements of the human silhouettes in frame. We also carried out performance analysis based on run time and time performance of algorithms. A fuzzy based analysis was carried out. The results show that our algorithms give a better performance than existing ones. The system works under existing infrastructures and ISPs without any modification, so as to reduce the level of complexity of the system.

Keywords: Segmentation algorithms, Surveillance systems, Anthropometry, Object Identification, Object classification and Face localization.