Improving Electron Micrograph Signal-to-Noise with an Atrous Convolutional Encoder-Decoder

Abstract

We present an atrous convolutional encoder-decoder trained to denoise 512$\times$512 crops from electron micrographs. It consists of a modified Xception backbone, atrous convoltional spatial pyramid pooling module and a multi-stage decoder. Our neural network was trained end-to-end to remove Poisson noise applied to low-dose ($\ll$ 300 counts ppx) micrographs created from a new dataset of 17267 2048$\times$2048 high-dose ($>$ 2500 counts ppx) micrographs and then fine-tuned for ordinary doses (200-2500 counts ppx). Its performance is benchmarked against bilateral, non-local means, total variation, wavelet, Wiener and other restoration methods with their default parameters. Our network outperforms their best mean squared error and structural similarity index performances by 24.6% and 9.6% for low doses and by 43.7% and 5.5% for ordinary doses. In both cases, our network's mean squared error has the lowest variance. Source code and links to our new high-quality dataset and trained network have been made publicly available at https://github.com/Jeffrey-Ede/Electron-Micrograph-Denoiser

arXiv comment15 pages, 10 figures, 1 table

References in corpus (29)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Deep Learning in Neural Networks: An Overview
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• TensorFlow: A system for large-scale machine learning
• Rethinking Atrous Convolution for Semantic Image Segmentation
• scikit-image: Image processing in Python
• Playing Atari with Deep Reinforcement Learning
• Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss
• A guide to convolution arithmetic for deep learning
• cuDNN: Efficient Primitives for Deep Learning
• Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
• A Review of Convolutional Neural Networks for Inverse Problems in Imaging
• Revisiting Distributed Synchronous SGD
• Achieving Human Parity in Conversational Speech Recognition
• Ultrafast transmission electron microscopy using a laser-driven field emitter: femtosecond resolution with a high coherence electron beam
• PlaNet - Photo Geolocation with Convolutional Neural Networks
• Why does deep and cheap learning work so well?
• Comparison of optimal performance at 300 keV of three direct electron detectors for use in low dose electron microscopy
• Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections
• Batch Renormalization: Towards Reducing Minibatch Dependence in Batch-Normalized Models
• Superhuman Accuracy on the SNEMI3D Connectomics Challenge
• Regularization for Deep Learning: A Taxonomy
• A deep convolutional neural network approach to single-particle recognition in cryo-electron microscopy
• A Deep Convolutional Neural Network to Analyze Position Averaged Convergent Beam Electron Diffraction Patterns
• Improving Electron Micrograph Signal-to-Noise with an Atrous Convolutional Encoder-Decoder
• Deep Convolutional Denoising of Low-Light Images
• Beating the World's Best at Super Smash Bros. with Deep Reinforcement Learning

Cited by in corpus (9)

• Review: Deep Learning in Electron Microscopy
• Leveraging generative adversarial networks to create realistic scanning transmission electron microscopy images
• Developing and Evaluating Deep Neural Network-based Denoising for Nanoparticle TEM Images with Ultra-low Signal-to-Noise
• Partial Scanning Transmission Electron Microscopy with Deep Learning
• Improving Electron Micrograph Signal-to-Noise with an Atrous Convolutional Encoder-Decoder
• Warwick Electron Microscopy Datasets
• Optimal STEM Convergence Angle Selection using a Convolutional Neural Network and the Strehl Ratio
• Noise2Void for Denoising Atomic Resolution Scanning Transmission Electron Microscopy Images
• Convolutional neural networks applied to differential dynamic microscopy reduces noise when quantifying heterogeneous dynamics