ISAR target recognition based on manifold learning

摘要

In this paper, the idea of manifold learning is introduced into inverse synthetic aperture radar (ISAR) target recognition, a new method based on Locality Preserving Projections (LPP) algorithm and k-nearest neighbour classification for ISAR target recognition is proposed. Firstly, the LPP algorithm is used to reduce the dimension of ISAR image, and then three kinds of aircraft target are classified by k-nearest neighbour classification in the low-dimensional subspace. The simulated experimental results suggest that the proposed method has the capability of finding the low-dimensional manifold structure embedded in the high-dimensional ISAR image space controlled by few parameters, such as attitude angle, scale and position, etc., and better classification performance is acquired comparing to PCA and LDA.