SAR image segmentation based on analysing similarity with clutter spatial patterns

An image segmentation method is proposed based on exploiting the spatial patterns in synthetic aperture radar (SAR) images. First, a similarity map is constructed for evaluating the similarity between the spatial pattern of the SAR image and the reference clutter images. Then, a segmentation threshold is selected based on the similarity map. Finally, image segmentation is achieved by preserving pixels that have larger similarity values. The superiority of the proposed method is verified by the experimental results for high-resolution SAR images.

[1]  Huanxin Zou,et al.  A Bilateral CFAR Algorithm for Ship Detection in SAR Images , 2015, IEEE Geoscience and Remote Sensing Letters.

[2]  Hongming Zhou,et al.  Extreme Learning Machine for Regression and Multiclass Classification , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[3]  Fang Liu,et al.  Spectral Clustering Ensemble Applied to SAR Image Segmentation , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Zhipeng Liu,et al.  Adaptive boosting for SAR automatic target recognition , 2007, IEEE Transactions on Aerospace and Electronic Systems.

[5]  U. Leloglu,et al.  Segmentation of SAR images using similarity ratios for generating and clustering superpixels , 2016 .

[6]  Jun Zhang,et al.  A segmentation algorithm for SAR images based on the anisotropic heat diffusion equation , 2008, Pattern Recognit..