A target detection algorithm based on peripheral anomaly

Automatic target detection algorithms enable in military and civilian applications the detection of low-contrast and spatially small targets that can not be detected by a human operator and remove the requirement of permanent operator control for surveillance systems. Considering many different background patterns and target models, it is not possible to define a single target detection algorithm that succeeds in all scenarios. In this paper, a novel target detection algorithm is proposed for targets above sea-level and sky targets. Peripheral anomaly values are calculated for image blocks depending on the difference of the statistical data between target and background patterns and target and background regions are differentiated by comparing these values with scenario dependent threshold values. Experimental results show that the proposed method has high success rate in the detection of targets above sea-level and sky targets in visual band images.