Comparison of two methods for automatic range alignment in ISAR imaging

Abstract

A maximum likelihood algorithm for translational motion estimation (TME) in ISAR imaging is described and compared with prominent point processing (PPP). It is suggested that the kurtosis (fourth order central moment) of averaged range profiles can be used to decide on which of the two methods that gives the best results in terms of image sharpness. Experimental data of a small aircraft is collected and processed in short intervals. Results from the PPP and the TME algorithms are compared, automatically in terms of a sharpness measure, and also by user inspection. The results indicate that for low kurtosis regions the TME method should be used, while for high kurtosis regions, the PPP method is preferred.