Target automatic recognition based on ISAR image with wavelet transform and MBLBP

A novel feature extraction algorithm of Inverse Synthetic Aperture Radar (ISAR) image based on wavelet transform (WT) and multi-scale block local binary pattern (MB-LBP) is proposed in this paper. Firstly, the mathematical morphology method is adopted to enhance ISAR image. Secondly, 2D wavelet transform is used to get the low frequency sub-band image from the enhancement ISAR image. Then, ISAR image is divided into small regions from which block local binary pattern (BLBP) with different weight histograms are extracted and concatenated into a feature vector to be used as an efficient ISAR descriptor. At last the classification is performed by using a nearest neighbor classifier with Chi square as a dissimilarity measure in the computed feature space. The experimental results on three kinds of aircraft targets show that the method performs very well.