Microscopy Image Restoration using Deep Learning on W2S
This work addresses image restoration for biomedical microscopy, but it is incremental as it builds directly on an existing method (W2S) with minor variations.
The paper tackles the problem of jointly denoising and super-resolving biomedical fluorescence microscopy images, specifically restoring SIM images from widefield inputs, resulting in a model that increases resolution by a factor of two and provides visually-convincing denoising with inference times under 1 second on a GPU.
We leverage deep learning techniques to jointly denoise and super-resolve biomedical images acquired with fluorescence microscopy. We develop a deep learning algorithm based on the networks and method described in the recent W2S paper to solve a joint denoising and super-resolution problem. Specifically, we address the restoration of SIM images from widefield images. Our TensorFlow model is trained on the W2S dataset of cell images and is made accessible online in this repository: https://github.com/mchatton/w2s-tensorflow. On test images, the model shows a visually-convincing denoising and increases the resolution by a factor of two compared to the input image. For a 512 $\times$ 512 image, the inference takes less than 1 second on a Titan X GPU and about 15 seconds on a common CPU. We further present the results of different variations of losses used in training.