The only thing I can think of is the subset before the deburst. Maybe this causes the error regarding the “image’s bounds”.
@khaldi: Instead of subsetting your interferogram, select a smaller part already in the TOPS coregistration by reducing the number of bursts:
