A Focused Target Segmentation Paradigm

Abstract

In this paper we present new algorithms for target detection/segmentation in second generation Forward Looking Infra-Red (FLIR) images. An initial detection algorithm that models the background using Weibull functions, is used to identify candidate target locations in the image. A two-stage focused analysis of each candidate target location is then performed to get an accurate representation of the target boundary. A region-growing procedure is used to get an initial estimate of the target region, which is then combined with salient edge information in the image to arrive at a more accurate representation of the target boundary. The region and edge integration is done using a novel method that uses a Bayes' minimum risk classification approach. Finally, to reduce the false alarm rate, a higher level interpretation module is used to classify the detected areas as man-made or natural objects using geometric and FLIR-intensity based features extracted from the target.