Enhancing Water Resource Management: A GUI-based Approach for Water Body Classification and Change Detection in VHRS Images

The accurate classification of water bodies is crucial for effective water resource management, particularly when leveraging remote sensing and deep learning techniques. However, achieving precise classification in Very HighResolution Satellite (VHRS) images poses a significant challenge, necessitating accurate identification and categorization methodologies. Existing methodologies lack userfriendly interfaces and struggle with real-time integration into Geographic Information System (GIS) maps. This proposed system aims to develop a graphical user interface (GUI) that facilitates the classification of water bodies in VHRS images. The GUI employs preprocessing techniques such as Bilateral filtering and false color composites. Water body segmentation is performed using the U-Net model, followed by classification with the Random Forest Classifier. Additionally, the GUI facilitates change detection analysis, enabling to generate suitable vector data from raster images for the identification of temporal variations in water bodies. The detected changes are seamlessly integrated into GIS maps, ensuring the timely and accurate updating of spatial data. The proposed approach is evaluated using an urban dataset from Kolkata, West Bengal, India.