When computing stats for rescale=1 ==> perc_diff=0. This is not a normal behavior. When image is transformed it is re-segmented this not the case when rescale=1 which explains perc_diff=0.
Image must be segmented twice (for rescale 1)to account for the variation of CSA due to image transformation. Name one segmentation "gt" and the other "1.0"
When computing stats for rescale=1 ==> perc_diff=0. This is not a normal behavior. When image is transformed it is re-segmented this not the case when rescale=1 which explains perc_diff=0. Image must be segmented twice (for rescale 1)to account for the variation of CSA due to image transformation. Name one segmentation "gt" and the other "1.0"