We propose an effective deep learning model to denoise scanning transmission electron microscopy (STEM) image series, named Noise2Atom, to map images from a source domain S $\mathcal {S}$ to a target domain C $\mathcal {C}$ , where S $\mathcal {S}$ is for our noisy experimental dataset, and C $\mathcal {C}$ is for the desired clear atomic images. Noise2Atom uses two external networks to apply additional constraints from the domain knowledge. This model requires no signal prior, no noise model estimation, and no paired training images. The only assumption is that the inputs are acquired with identical experimental configurations. To evaluate the restoration performance of our model, as it is impossible to obtain ground truth for our experimental dataset, we propose consecutive structural similarity (CSS) for image quality assessment, based on the fact that the structures remain much the same as the previous frame(s) within small scan intervals. We demonstrate the superiority of our model by providing evaluation in terms of CSS and visual quality on different experimental datasets.

中文翻译：

---

Noise2Atom：用于扫描透射电子显微镜图像的无监督去噪

我们提出了一种有效的深度学习模型来去噪扫描透射电子显微镜 (STEM) 图像系列，名为 Noise2Atom，将图像从源域 S $\mathcal {S}$ 映射到目标域 C $\mathcal {C}$ ，其中 S $\mathcal {S}$ 用于我们嘈杂的实验数据集，而 C $\mathcal {C}$ 用于所需的清晰原子图像。Noise2Atom 使用两个外部网络来应用领域知识中的附加约束。该模型不需要信号先验，不需要噪声模型估计，也不需要成对的训练图像。唯一的假设是输入是通过相同的实验配置获得的。为了评估我们模型的恢复性能，因为我们的实验数据集不可能获得地面实况，我们提出了连续结构相似性 (CSS) 用于图像质量评估，基于这样一个事实，即结构在小扫描间隔内保持与前一帧大致相同。我们通过在不同实验数据集上提供 CSS 和视觉质量方面的评估来证明我们模型的优越性。