Fig. 1
From: Convolutional neural network for automated mass segmentation in mammography
The U-Net architecture consists of convolutional encoding and decoding units that take an image as input and produce the segmentation feature maps with respective pixel classes. The yellow arrows show the skip connections between the layers