This is a re-implementation of the Orobix's retina-unet model for retina blood vessel segmentation using Tensorflow rather than Keras. In addition, all pre-processing was implemented from scratch, with the exception of the CLAHE method. All code was implemented using the Orobix project only as a reference. This project is for academic purposes only.
Cortinovis, Daniele 2016. Retina blood vessel segmentation with a convolution neural network (U-net). https://github.com/orobix/retina-unet.git