A novel technique for feature-based aircraft identification from high resolution airborne ISAR images

In this paper we propose a new method for aircraft identification and classification using high-resolution ISAR images acquired from an airborne radar sensor. The proposed method takes advantage of some of the geometric features extracted from the image, and some signal features as well (like the Jet Engine Modulation phenomenon that allows us detecting and analyzing the engines of the unknown aircraft). In the next session we will provide a brief explanation of the problem, in order to show the proposed method. An ad hoc classification database has been developed, containing physical characteristics of some aircrafts. Features extracted from radar data have been compared with the database content for classification and some preliminary results have already been obtained. These results appear encouraging and will be shown in the paper.