Adaptive Fuzzy Learning Superpixel Representation for PolSAR Image Classification

The increasing applications of polarimetric synthetic aperture radar (PolSAR) image classification demand for effective superpixels’ algorithms. Fuzzy algorithms reduce the misclassification rate by dividing pixels into superpixels, which are groups homogenous appearance and undetermined pixels. However, two key issues remain to be addressed in designing a fuzzy superpixel algorithm PolSAR classification. First, scattering information, is unique images, not effectively used. Such information can utilized generate superpixels more suitable images. Second, ratio fixed each existing techniques, ignoring fact that difficulty classifying different objects varies an image. To address these issues, we propose information-based adaptive (AFS) images In AFS, correlation between pixels’ first time, considered through rough set theory superpixels. This further used dynamically adaptively update AFS evaluated extensively against evaluation metrics compared with state-of-the-art on three experimental results demonstrate superiority problems.