UNet
The U-Net is convolutional network architecture for fast and precise segmentation of images. Up to now it has outperformed the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks.
The provided model is basically a convolutional auto-encoder, but with a twist - it has skip connections from encoder layers to decoder layers that are on the same "level". See picture below (note that image size and numbers of convolutional filters in this tutorial differs from the original U-Net architecture).
This deep neural network is implemented with Keras functional API, which makes it extremely easy to experiment with different interesting architectures.
Output from the network is a 64 x 80 which represents mask that should be learned. Sigmoid activation function makes sure that mask pixels are in [0, 1] range.
Our U-Net implementation is based on an initial implementation by raun1 although both network and parameters has been redesigned to match the U-Net paper's architecture as well as optimizations applied.
Ronneberger, O., Fischer, P., & Brox, T. (2015, October). U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention (pp. 234-241). Springer, Cham.