Target recognition of SAR images by partially matching of target outlines

A synthetic aperture radar (SAR) automatic target recognition method is proposed based on the matching of the target outlines. The target outline describes the physical sizes and shape of the target thus discriminative for SAR target recognition. The original target outline is segmented into several independent parts. The distance between each part and its counterpart in the corresponding template is measured by the least-trimmed square Hausdorff distance. Afterwards, the results of individual parts are combined to form a similarity measure, which comprehensively considers the possible deformations of the target outline. Based on the similarity measure, the target type is determined to be the class sharing the maximum similarity with the test sample. To evaluate the performance of the proposed method, extensive experiments are conducted on the Moving and Stationary Target Acquisition and Recognition dataset under both the standard operating condition and several typical extended operating conditions.