Image denoising using deep neural networks
keywords DEEP LEARNING, VIDEO ANALYSIS
Reference persons ENRICO MAGLI
Research Groups CCNE - COMMUNICATIONS AND COMPUTER NETWORKS ENGINEERING, ICT4SS - ICT FOR SMART SOCIETIES, Image Processing Lab (IPL)
Thesis type RESEARCH
Description Images are always affected by noise and this is particularly true for photos taken in low-light conditions or astronomical images. Denoising algorithms are crucial to remove the undesired noise and recover the
clean image. Recently, neural networks have shown promising results for this problem. In this thesis, the candidate will design neural networks for different kinds of noise using cutting-edge methods like adversarial
learning, graph-based convolution, and residual networks. The architecture can be specialized to target different kinds of noise such as Gaussian noise in natural photos, Poisson noise in MRI and few-photon imaging, speckle in synthetic aperture radar imaging.
Required skills Candidate students should have a background in ICT or mathematics. Knowledge of neural networks (including TensorFlow environment and Python programming) is not a prerequisite, although it would help in the initial stage of the activity.
Deadline 27/09/2019 PROPONI LA TUA CANDIDATURA