Deep Denoising for Scientific Discovery: A Case Study in Electron Microscopy
Deep Denoising for Scientific Discovery: A Case Study in Electron Microscopy
复制标题
DOI：
10.1109/tci.2022.3176536
复制
发表时间：
2022-01-01
影响因子：
5.4
通讯作者：
Fernandez-Granda, Carlos
中科院分区：
文献类型：
--
作者：
Mohan, Sreyas;Manzorro, Ramon;Fernandez-Granda, Carlos
去噪是科学成像中的一个基本挑战。深度卷积神经网络（CNN）提供了对摄影图像进行去噪的最新技术水平。然而，它们的潜力在科学成像方面还没有得到充分的探索。去噪CNN通常是在被人工噪声污染的干净图像上训练的，但在科学应用中，无噪声的地面实况图像通常不可用。为了解决这个问题，我们提出了一个基于模拟的去噪（SBD）框架，其中CNN在模拟图像上进行训练。我们测试的框架上的透射电子显微镜（TEM）的数据，表明它优于现有的技术在模拟基准数据集，和真实的数据。我们分析了SBD的泛化能力，表明训练的网络对成像参数和底层信号结构的变化具有鲁棒性。我们的研究结果表明，用于对摄影图像进行去噪的最先进的架构可能无法很好地适应科学成像数据。例如，大幅增加它们的视场可以显著提高它们在低信噪比下获得的TEM图像上的性能。我们还表明，标准的性能指标的照片（如峰值信噪比）可能没有科学意义，并提出了几个指标来解决这个问题的TEM图像的情况下。此外，我们提出了一种技术，基于似然计算，可视化的协议之间的结构去噪图像和观察到的数据。最后，我们发布了一个公开的基准数据集，包含18，000个模拟TEM图像。
Denoising is a fundamental challenge in scientific imaging. Deep convolutional neural networks (CNNs) provide the current state of the art in denoising photographic images. However, their potential has been inadequately explored for scientific imaging. Denoising CNNs are typically trained on clean images corrupted with artificial noise, but in scientific applications, noiseless ground-truth images are usually not available. To address this, we propose a simulation-based denoising (SBD) framework, in which CNNs are trained on simulated images. We test the framework on transmission electron microscopy (TEM) data, showing that it outperforms existing techniques on a simulated benchmark dataset, and on real data. We analyze the generalization capability of SBD, demonstrating that the trained networks are robust to variations of imaging parameters and of the underlying signal structure. Our results reveal that state-of-the-art architectures for denoising photographic images may not be well adapted to scientific-imaging data. For instance, substantially increasing their field-of-view dramatically improves their performance on TEM images acquired at low signal-to-noise ratios. We also demonstrate that standard performance metrics for photographs (such as peak signal-to-noise ratio) may not be scientifically meaningful, and propose several metrics to remedy this issue in the case of TEM images. In addition, we propose a technique, based on likelihood computations, to visualize the agreement between the structure of the denoised images and the observed data. Finally, we release a publicly available benchmark dataset containing 18,000 simulated TEM images.