Klassifisering av skip ved bruk av ISAR

Abstract

ISAR (Inverse Synthetic Aperture Radar) is a radar mode for high-resolution imaging of moving targets. Airborne ISAR has a potential for long-range ship classification. This report presents a method for automatic ship classification. The theoretical basis is examined and an implemented algorithm is described. Our approach to automatic classification is based on 3-dimensional wire-frame models. The models are built from photographs using the commercial software PhotoModeler. The classification algorithm includes the following steps: The ISAR image is analysed, and specific features are extracted. The results are used for transforming the wire-frame models to the ISAR image projection plane. The silhouettes of the ISAR image and the transformed models are compared, and the best match is presented to the user.