Retinal Vessel and Cell Nuclei Segmentation
In this repository, segmentation is applied using u-net network for two data sets.
The U-net network has been trained using google colab. Repository includes code and data set.
Images in the retinal vessels dataset are cropped to 512 x 512.
Images in the data set containing cell images are resized to 256 x 256.
Ground truth images in the cell data set were prepared as multiple files, these image files were turned into a single image.