Review on Machine Learning-based Change Detection and Pan Sharpening Algorithms for Remote Sensing Datasets

The Himalayan area of India experiences the most snowfall. Monitoring snow cover in the area is crucial for managing water supplies, performing hydrological research, and forecasting avalanches via remote sensing. Pan sharpening, a fusion technique that combines high-resolution and lower-resolution images, has proven to be an effective approach for enhancing spatial resolution and facilitating accurate snow cover change detection. This study focuses on providing a comprehensive analysis of the methodologies and advancements in snow cover change detection using pan sharpening techniques. The research is divided into three main sections: classification, change detection, and pan-sharpening. Classification methods are categorized as supervised and unsupervised, with applications ranging from image classification to land cover mapping. Change detection is critical for identifying environmental changes over time, and various techniques like change vector analysis, artificial neural networks, and post classification comparison are discussed. Pan-sharpening techniques, including deep learning-based and traditional methods, are explored for enhancing image resolution and spectral detail. We have reviewed the key concepts, challenges, and recent developments in pan-sharpening-based snow cover change detection, highlighting the strengths and limitations of different techniques.