Bayesian change analysis for finding vehicle size targets in VHF foliage penetration SAR data

Abstract

A Bayesian change analysis detection scheme for VHF SAR data is presented. It is notably different from the conventional approach of tresholding target probability of detection and false alarm rate. The approach leads to target hypotheses, in which target nominees are attributed with a probability of being a target or equivalently of not being a false alarm. The precise method is iterative, substituting each hypothesis with a new one containing one further target. It stops when the probability of further targets decreases monotonically. In typical situations it peaks with a very high probability for a certain number of targets, which thus is the most likely distribution of targets given the image pair.