SAR image segmentation using morphological thresholding

In this paper, a new approach is proposed for segmentation of synthetic aperture radar (SAR) images. This method is based on image filtering and thresholding using morphological operations. Image segmentation is an important step for remote sensing applications. Clustering methods are one kind of the common techniques for segmentation; however, these techniques have high computational complexity and their results, especially in the field of SAR image segmentation, are not appropriate. Among the above methods, the Kmeans clustering and spectral clustering (SC) can be noted. This work presents an innovative way in comparison with previous clustering methods. Experimental results show that the proposed method has less error percentage and it is more accurate for segmentation of SAR images.