SOTA	state of the art
BA2Net	boundary-aware attentive network
CM	change map
CCM	coarse change map
RCM	refined change map
VHR	very high resolution
SAR	synthetic aperture radar
CNN	convolutional neural network
DBN	deep brief network
SDAE	sparse de-noising autoEncoder
FCN	fully convolutional network
BN	batch normalize
ReLU	rectified linear unit
AG	attention gate
GT	ground truth
TP	true positive
TN	true negative
FP	false positive
FN	false negative
OA	overall accuracy
CE	cross entropy
BCE	binary cross entropy
SSIM	structural similarity
HVS	human visual system
CD-Net	change detection network
FC-EF	fully convolutional early fusion
FC-Siam-conc	fully convolutional Siamese concatenation
FC-Siam-diff	fully convolutional Siamese difference
FCN-PP	fully convolutional network with pyramid pooling
DSCN	deep Siamese convolutional network
DSMS	deep Siamese multiScale
MSOF	multiple side-outputs fusion
IFN	image fusion network