Weak-signal extraction enabled by deep-neural-network denoising of diffraction data
This addresses the challenge of accurately denoising data for scientific applications where ground truth fidelity is crucial, though it is incremental as it applies an existing deep learning method to a specific domain problem.
The researchers tackled the problem of extracting weak signals from noisy scientific data, specifically X-ray diffraction on crystalline materials, by using a deep convolutional neural network for denoising, which made previously insignificant charge ordering signals visible and quantitatively accurate.
Removal or cancellation of noise has wide-spread applications for imaging and acoustics. In every-day-life applications, denoising may even include generative aspects, which are unfaithful to the ground truth. For scientific use, however, denoising must reproduce the ground truth accurately. Here, we show how data can be denoised via a deep convolutional neural network such that weak signals appear with quantitative accuracy. In particular, we study X-ray diffraction on crystalline materials. We demonstrate that weak signals stemming from charge ordering, insignificant in the noisy data, become visible and accurate in the denoised data. This success is enabled by supervised training of a deep neural network with pairs of measured low- and high-noise data. We demonstrate that using artificial noise does not yield such quantitatively accurate results. Our approach thus illustrates a practical strategy for noise filtering that can be applied to challenging acquisition problems.