1. INTRODUCTION
Polarimetric Synthetic Aperture Radar (PolSAR) data have been utilized to identify cumulative changes in land cover by comparing and analyzing a time series of images covering the same geographic area at different times. In the past, people typically employed methods such as the generalized maximum likelihood ratio test (MLR)[1], the complex-kind Hotelling–Lawley trace (HLT)[2], the Symmetric Revised Wishart Distance (SRWD)[3], or the determinant ratio test (DRT) statistic [4] for paired change detection using images from two consecutive adjacent time points in a time series. Subsequently, all the change detection results were fused to obtain a cumulative change binary map of the entire time series. This approach required multiple paired change detections, each involving the difference image (DI) calculation and DI segmentation. The repetitive calculations led to low efficiency in the overall processing. Additionally, the need for thresholding or clustering in each paired change detection introduced the possibility of omissions or errors in cumulative change detection results, particularly for small changes at specific points in time.