drusen-segmentation

This is the implementation code for paper submitted to Bioinformatics: "Unsupervised deep feature extraction via adaptive collaborative learning for drusen segmentation of fundus images".

required environment

C++, matlab

Datasets

STARE dataset available: http://cecas.clemson.edu/~ahoover/stare/

DRIVE dataset available: https://www.isi.uu.nl/Research/Databases/DRIVE/

Code

1. To train deep_net and SVM classifier based on train data:

   main_train.m (drusen_segmentation_demo.m; training.m; net_struc,m; updating_net.m)

2. To test the model based on test data:

   main_test.m (test_patches.m; test_evaluation.m)