Skip to content

Latest commit

 

History

History
44 lines (29 loc) · 956 Bytes

README.md

File metadata and controls

44 lines (29 loc) · 956 Bytes

Feature based loss function for retinal multi-class vessel segmentation and feature measurement

To do list

Add more description and binary vessel segmentation.

Pretrained Model

The pretrained models in Google_DRIVE. Download them and unzip them.

Train

Start training, the dataset can be set as DRIVE_AV, LES-AV, or HRF-AV.

python train.py --e=1000 \
                --batch-size=2 \
                --learning-rate=8e-4 \
                --v=10.0 \
                --alpha=0.5 \
                --beta=1.0 \
                --gama=1.0 \
                --dataset='DRIVE_AV' \
                --loss='CF' \
                --uniform=True

Test

Test the trained models.

python test.py --batch-size=1 \
                --uniform=True \
                --test_dataset='DRIVE_AV' \
                --loss='CF'

Citation

TBC