Spatial Fuzzy Clustering on Synthetic Aperture Radar Images to Detect Changes

TL;DR

This paper introduces a spatial fuzzy clustering method for synthetic aperture radar images to effectively detect changes and mitigate speckle noise, enhancing accuracy in environmental and urban studies.

Contribution

It proposes a novel spatial fuzzy clustering approach that leverages spatial information to improve change detection in SAR images, addressing speckle noise challenges.

Findings

Enhanced clustering accuracy in SAR images

Effective speckle noise reduction

Improved change detection performance

Abstract

Data and data sources have become increasingly essential in recent decades. Scientists and researchers require more data to deploy AI approaches as the field continues to improve. In recent years, the rapid technological advancements have had a significant impact on human existence. One major field for collecting data is satellite technology. With the fast development of various satellite sensor equipment, synthetic aperture radar (SAR) images have become an important source of data for a variety of research subjects, including environmental studies, urban studies, coastal extraction, water sources, etc. Change detection and coastline detection are both achieved using SAR pictures. However, speckle noise is a major problem in SAR imaging. Several solutions have been offered to address this issue. One solution is to expose SAR images to spatial fuzzy clustering. Another solution is to separate speech. This study utilises the spatial function to overcome speckle noise and cluster the SAR images with the highest achieved accuracy. The spatial function is proposed in this work since the likelihood of data falling into one cluster is what this function is all about. When the spatial function is employed to cluster data in fuzzy logic, the clustering outcomes improve. The proposed clustering technique is us