This repository is the official PyTorch code for the paper 'Full-Resolution Network and Dual-Threshold Iteration for Retinal Vessel and Coronary Angiograph Segmentation'.
Download our repo:
git clone https://github.com/lseventeen/RF-UNet.git
cd RF-UNet
Install packages from requirements.txt
pip install -r requirements.txt
Choose a path to create a folder with the dataset name and download datasets DRIVE,CHASEDB1,STARE,CHUAC, and DCA1. Type this in terminal to run the data_process.py file
python data_process.py -dp DATASET_PATH -dn DATASET_NAME
Type this in terminal to run the train.py file
python train.py -dp DATASET_PATH
Type this in terminal to run the test.py file
python test.py -dp DATASET_PATH -wp WEIGHT_FILE_PATH
We have prepared the pre-trained models for both datasets in the folder 'pretrained_weights'. To replicate the results in the paper, directly run the following commands
python test.py -dp DATASET_PATH -wp pretrained_weights/DATASET_NAME
This project is licensed under the MIT License - see the LICENSE file for details
CUDA_VISIBLE_DEVICES=1 python train.py -dp "/home/xjiangbh/Retina/Data/DRIVE/" -dn DRIVE --val CUDA_VISIBLE_DEVICES=0 python train.py -dp "/home/xjiangbh/Retina/Data/CHASEDB1/" -dn CHASEDB1 --val CUDA_VISIBLE_DEVICES=2 python train.py -dp "/home/xjiangbh/Retina/Data/STARE/" -dn STARE --val