Figure 8
From: Semantic segmentation of PolSAR image data using advanced deep learning model
The modified Xception architecture used in DeepLabv3+: (1) has higher number of layers, (2) all max pooling layers are replaced by depthwise seperable convolutions with striding and (3) extra batch normalization and ReLU are added after each 3 × 3 depthwise convolution42.
