Materials Data Science and Informatics (IAS-9)

In this project, we leverage deep learning and computer vision methods to address key challenges in transmission electron microscopy (TEM) image processing. The primary objective is to develop techniques for denoising, background removal, and super-resolution in TEM images, enhancing their quality and enabling more accurate analysis. By harnessing the power of deep neural networks, we aim to effectively suppress noise, remove unwanted background artifacts, and enhance spatial resolution, thereby improving the clarity and interpretability of TEM images. This provides researchers with invaluable tools for uncovering subtle features and structures at the nanoscale. This project potentially facilitates progress in materials science, nanotechnology, and various other domains reliant on precise imaging and analysis of nanomaterials. In this project, we use the publicly available TEMImageNet dataset (Lin et al., 2021).