A Decision Fusion Framework for ISAR Images Recognition Based On Scattering Center and Zernike Moment

Pengling Tang, Bo Long, Feng Wang

Abstract: Under the condition of large difference in imaging angle, Inverse Synthetic Aperture Radar (ISAR) images vary greatly and are difficult to identify. To solve this problem, a decision fusion framework based on scattering centers and Zernike Moment is proposed. In this framework, each feature corresponds to a classifier, then these classifiers are combined together by assigning different weights which reflect the ability of the classifiers. Experiments conducted on simulated ISAR images demonstrate that the proposed method can combine the outputs of the classifiers effectively and achieve better performance than single classifier.