Deep neural networks for understanding noisy data applied to physical property extraction in scanning probe microscopy
Deep neural networks for understanding noisy data applied to physical property extraction in scanning probe microscopy
复制标题
DOI：
10.1038/s41524-019-0148-5
复制
发表时间：
2019-02-22
影响因子：
9.7
通讯作者：
Jesse, Stephen
中科院分区：
文献类型：
--
作者：
Borodinov, Nikolay;Neumayer, Sabine;Jesse, Stephen
过去十年中，扫描探针、电子和光学显微镜等光谱成像方法的快速发展催生了大型多维数据集。在许多情况下，将高光谱数据简化为低维材料特定参数是基于函数拟合，其中拟合函数的近似形式是已知的，但需要确定函数的参数。然而，通过迭代方法（例如最小二乘梯度下降）实现的噪声数据的函数拟合通常会产生虚假结果，并且对初始猜测非常敏感。在这里，我们演示了一种使用深度神经网络方法减少高光谱数据的方法。深度神经网络/最小二乘相结合的方法被证明可以将带激励压电响应力显微镜的有效信噪比提高一个数量级以上，从而可以在使用非常小的驱动信号或材料响应较弱时进行表征。
The rapid development of spectral-imaging methods in scanning probe, electron, and optical microscopy in the last decade have given rise for large multidimensional datasets. In many cases, the reduction of hyperspectral data to the lower-dimension materials-specific parameters is based on functional fitting, where an approximate form of the fitting function is known, but the parameters of the function need to be determined. However, functional fits of noisy data realized via iterative methods, such as least-square gradient descent, often yield spurious results and are very sensitive to initial guesses. Here, we demonstrate an approach for the reduction of the hyperspectral data using a deep neural network approach. A combined deep neural network/least-square approach is shown to improve the effective signal-to-noise ratio of band-excitation piezoresponse force microscopy by more than an order of magnitude, allowing characterization when very small driving signals are used or when a material's response is weak.