The ResUNet++ architecture is based on the Deep Residual U-Net (ResUNet), which is an architecture that uses the strength of deep residual learning and U-Net. The proposed ResUNet++ architecture takes advantage of the residual blocks, the squeeze and excitation block, ASPP, and the attention block.
ResUNet++: An Advanced Architecture for Medical
Image Segmentation
The following datasets are used in this experiment:
- Batch size = 16
- Number of epoch = 300
- Loss = Binary crossentropy
- Optimizer = Nadam
- Learning Rate = 1e-5 (Adjusted for some experiments)
Qualitative result comparison of the proposed models with UNet, ResUNet, and ResUNet++.