Change detection in remote sensing refers to the process of identifying differences in the state of an object or phenomenon by observing it at different times. It is a critical technique used in environmental monitoring, urban planning, agriculture, forestry, and disaster management. The primary goal of change detection is to quantify, locate, and analyze the changes that occur over time in the Earth's surface by comparing multi-temporal satellite or aerial images. These changes can be natural, such as vegetation growth, soil erosion, or water level fluctuations, or anthropogenic, such as urban expansion, deforestation, and land use changes. Accurate change detection requires high-quality, pre-processed images that are geometrically aligned, radiometrically corrected, and temporally consistent to minimize errors caused by differences in sensor calibration, atmospheric conditions, and seasonal variations. Classification-based methods of change detection involve categorizing land cover types in multi-temporal images and then comparing these classifications to identify changes. These methods can be broadly divided into two approaches: post-classification comparison and direct multi-date classification. In post-classification comparison, each image acquired at different times is independently classified using techniques such as supervised classification, unsupervised classification, or hybrid approaches. Supervised classification