ZS-Denoiser-HREM

This repository is Pytorch implementation of our manuscript "Zero-shot Image Denoising for High-Resolution Electron Microscopy" [ArXiv][IEEE Xplore].

Pipeline of ZS-Denoiser

Fig. 1: The pipeline of ZS-Denoiser HREM.

---

Simulated HREM Denoising Example

The simulated TEM dataset released by Mohan et al. [Github] which consists of approximate 18000 simulated images. Fig. 2: Comparison of denoising results of simulated Pt/CeO2 catalyst corrputed with Poisson-Gaussain noise.

Real HREM Denoising Example

Fig. 3 Comparison of denoising results of real STEM data on zeolites.

---

1. Running Environment

To run this project, you will need the following packages:

• Pytorch
• Scikit-image
• Tiffile, tqdm, numpy and other packages.

2. File Tree

ZS-Denoiser-HREM
│  dataset.py 
│  netarch.py             
│  README.md
│  train.py               # train zero-shot denoising network
│  utils.py
│          
├─config
│      simulated_PG.json  # configuration file
│      
└─demo_data
     └─PtCeO2_simulated   # simulated data for numerical experiments
            1.tif
            2.tif
            3.tif
            4.tif
            5.tif

3. Train the ZS-Denoiser on simulated HREM image

To train the denoising model for simulated HREM image corrupted with Poission-Gaussain noise ($a = 0.05, b = 0.02$), you can run the following command in your terminal:

python train.py -image_path demo_data/PtCeO2_simulated/1.tif -a 0.05 -b 0.02

4. License

This code is available for non-commercial research and education purposes only. It is not allowed to be reproduced, exchanged, sold, or used for profit.

5. Citation

If you find our work useful in your research, please site:

@ARTICLE{10675590,
  author={Tian, Xuanyu and Dong, Zhuoya and Lin, Xiyue and Gao, Yue and Wei, Hongjiang and Ma, Yanhang and Yu, Jingyi and Zhang, Yuyao},
  journal={IEEE Transactions on Computational Imaging}, 
  title={Zero-Shot Image Denoising for High-Resolution Electron Microscopy}, 
  year={2024},
  volume={10},
  number={},
  pages={1462-1475},
  keywords={Noise measurement;Noise reduction;Training;Image denoising;Signal to noise ratio;Electrons;Imaging;Denoising;electron microscopy;self-supervised;zero-shot},
  doi={10.1109/TCI.2024.3458411}}