This document discusses change detection using a hypothesis test-based approach. It describes how hypotheses are formulated as either a null hypothesis (H0) of no change or an alternative hypothesis (Hk) of a change at a given position k between images. A generalized likelihood ratio test is used to compute the difference in log-likelihood between the hypotheses. A threshold is then applied to determine if the difference is significant enough to indicate a true change. The approach aims to detect step changes over time in images for applications like monitoring natural disasters. Non-local filtering can also be used to increase the effective number of looks and improve detection performance.