Skip to content

Using physical and statistical principles of light to enhance denoising of cell images

Notifications You must be signed in to change notification settings

EmaadKhwaja/Deep-Learning-Denoising

Repository files navigation

Deep-Learning-Denoising

The focus of this project is using deep learning to denoise cellular images.

A modification of the Recurrent Inference Machine (RIM) architecture is the main technique employed.

Herein, we use the log likelihood of the Poisson distribution in order to successfully denoise cell images which are dominated by signal-dependent noise rather than additive gaussian white noise.

Much credit to Patrick Putzky for uploading some base code.

About

Using physical and statistical principles of light to enhance denoising of cell images

Topics

Resources

Stars

Watchers

Forks