search

orobix / retina-unet

Retina blood vessel segmentation with a convolutional neural network
1.26k stars 468 forks source link

Data Augmentation #12

Open ccccly opened 7 years ago

ccccly commented 7 years ago

Hi, I have some questions about the data augmentation. I have already finished reading the code in your package ,but I didn't find the code about random elastic deformations of the training samples. I found that the pathches were extracted from the images randomly. It's different from the paper: U-Net: Convolutional Networks for Biomedical Image Segmentation.

dcorti commented 7 years ago

Yes, you're right. I extracted the patches randomly, but I did not performed any data augmentation on these patches. Probably some data augmentation, like in the original U-net paper, could increase the performance but I have not tried yet. The results obtained were already quite satisfactory, so I left the data augmentation task as a possible future test. I hope to come back to it sooner or later :)