Aircraft Recognition Using ISAR Image Based on Quadrangle-points Affine Transform

Abstract

In radar automatic target recognition (RATR), inverse synthetic aperture radar (ISAR) image recognition shows its advantages. During ISAR image feature extraction, the feature points extraction is a classic method. However, the scatterer amplitude might appear to undulate, causing the failure of the matching pair, and affecting the result of recognition. To solve this problem, this paper proposes a quadrangle-points affine transform reconstruction (QATR) algorithm. Firstly, using the four structures of the aircraft nose, tail, and wings, the affine transform coefficients are calculated to reconstruct the aircraft as the top view. Then the template matching algorithm is adopted to recognize the target. The proposed method only needs few templates for each class and has robustness with attitudes sensitivity. The experiments based on real data demonstrate the effectiveness of this method.