Deep Learning

This part describes 3 applications we have studied with deep learning:

Steganography

You can see our paper entitled: Improving Blind Steganalysis in Spatial Domain using a Criterion to Choose the Appropriate Steganalyzer between CNN and SRM+EC

This paper has been accepted to  IFIP SEC 2017

Image denoising

Here are some examples of image denoising using CNN. In the following we give for each image three versions (noisy image, denoised image, and original image). Click on these images to see the normal size

Medical image segmentation

Semantic segmentation of myocardial infarction (MI) using MRI acquired several minutes after injection of a contrast agent. Our ISITE UBFC 2018 project has just been accepted. It is called: Automatic Detection of Viable myocArdiac segmeNts Considering dEep networkS (ADVANCES)