Automatic Classification of ISAR Images

Abstract

In this paper, we proposed a new classification method for the automatic classification of the aircraft targets from 2D-ISAR imagery. The proposed method depends on INVIgorated Rudimentary Silhouette (INVIRS) representation for improved target discrimination and classification. The INVIRS representation is more precise than existing shape matrix representations.  It is invariant to blur and small discontinuities in the superstructures of the targets. The performance of the proposed method is analyzed experimentally, and the results are presented and discussed.