Man-made Target Detection And Discrimination In SAR Imagery

Automatic target detection and recognition are important for civil and military applications, such as detecting the underground pipes, inspecting the tanks in the trees and ships on the sea, searching and rescuing the crashed aircraft, etc. Because Synthetic Aperture Radar (SAR) has the two primary advantages: all-weather and day or night imaging, target detection and recognition in SAR imagery will have wider applications in the future.A typical Automatic Target Recognition system consists of three stages: detection, discrimination and classification. Detection, whose role is to find regions in SAR imagery that contain potential targets, will inevitably produce false alarms. The false alarms are then further removed by the following stage: discrimination. Classification is to determine which class the target belongs to. This paper mainly discusses the problem about detection and discrimination, and contains the sections as follow:1) Research on the current situation in the field of automatic target detection and discrimination in SAR imagery: we present an overview on the algorithms and their advantages and drawbacks for automatic target detection and discrimination in SAR imagery. At present, there are three kinds of methods in the field of automatic target detection: CFAR, multi-resolution model, detection methods based on image phases or azimuth information, and the third one is the latest and most effective way among them, but it is not mature enough and requires further development. Target detection is considered as a process of two-class classification. It can be passed over if the false alarm is low enough.2) Comparison between ML-CFAR( Maximum Likelihood Constant False Alarm Rate, ML-CFAR) and 2L-IHP (Two-Look Internal Hermitian Product, 2L-IHP) algorithms: using the ADTS data set from Lincoln Laboratory, USA, we demonstrate that 2L-1HP can not only detect all...