Multiple human body tracking based on Normalized Cross-Correlation of Average Histogram using the Fusion of Color and thermal image sensor

Abstract

In this paper, a novel approach for tracking multiple human bodies is proposed by using sensor fusion with respect to color and thermal image sensors. In our human status monitoring system, a color and thermal image sensor is used together to make a robust human tracking. The Normalized Cross-Correlation of Average Histogram (NCCAH) Algorithm is used for analyzing between detected human bodies. Our strategy for detection and tracking is to analyze recursively the correlation between texture and thermal data. Our system tracks the moving human bodies by using the proposed NCCAH algorithm, which uses the static information of the detected bodies and the dynamic information of changes between previous and current frames. Our experiments show that the method yields good detection and tracking performance in dynamic environments, such as an occlusion, various electric heating appliances or complicated backgrounds.