ProgramMaster Logo
Conference Tools for Materials Science & Technology 2019
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools
About this Abstract
Meeting Materials Science & Technology 2019
Symposium Data Science for Material Property Interpretation
Presentation Title Deep Learning and MC-X ray, toward Automatic Sample Segmentation
Author(s) Samantha Rudinsky, Yu Yuan, Raynald Gauvin, Mike Marsh, Benjamin Provencher, Nicolas Piché
On-Site Speaker (Planned) Nicolas Piché
Abstract Scope Image segmentation, the process of labeling pixels according to their composition, is nearly always required to achieve proper quantitative analysis in materials science imaging studies. Deep Learning, the application of multi-layer neural networks, has recently emerged as the state-of-the-art method. Unfortunately, successful networks require extensive time-consuming human-curated segmentation examples that must serve as training data, and the resulting models fail to generalize beyond the highly specific images they were trained to solve. We present here a solution for simulating training data. We have integrated MC X-Ray simulation platform with our Deep Learning engine so that we can generate synthetic backscatter electron (BSE) images of artificial samples, sufficient for training neural networks that can later be used to segment real, and often noisy, BSE images. By scaling up to deeper networks with vastly more training data, our models show promise for fully automatic segmentation across a broader range of materials.
Proceedings Inclusion? Definite: At-meeting proceedings

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

3D Nanoprinting: An Integrated Approach of Experiments, Computer-aided Design and Simulations
4D STEM Data Acquisition, Analytics and Functional Material Property Extraction
A Machine Learning Approach to Independent Component Analysis for Nuclear Magnetic Resonance Spectra
Adversarial Networks for Digital Microstructure Generation
Application of a Statistical Analysis Technique for Characterizing the Deformation Behavior of the Material under Dynamic Impact Loading
Application of Artificial Neural Networks to Low Cycle Fatigue and Creep Data Processing for Power Plant Materials
Automated Defect Detection in Electron Microscopy with Machine Learning
Data Analytics for Correlative Multimodal Chemical and Functional Imaging
Data Science and the MGI
Deciphering the Atomic Origin of Glasses’ Properties by Machine Learning
Deep Learning and MC-X ray, toward Automatic Sample Segmentation
Digital Protocols for Statistical Quantification of Microstructure Features in Polycrystalline Nickel-based Superalloys
Human-in-the-loop Strategies for Dimensionality Reduction and Optimization in Materials Design
Model-based Reconstruction Algorithms for Time-of-Flight Neutron Tomography
Modeling and Simulation of Rare Events in Multidimensional Spaces
Multi-modal Data Fusion and 3D Reconstruction of Serial Sectioning Data
Neural Networks for Processing of Low Signal-to-noise Data in Scanning Probe Microscopy
P2-8: Evaluation for the Quality of Flake Graphite Cast Iron and Spheroidal Graphite Cast Iron by Tapping Test with Using Artificial Intelligence
Phase Field Regularization for Optimal Grain Reconstruction of Noisy Images
Python For Glass Genomics (PyGGi): A Machine Learning Package to Predict the Properties of Glasses
Recent Advances in 3D Reconstruction Based on Spherical Indexing of EBSD Data
Structure Prediction and Property-based Optimization of Molecular Crystals with GAtor
Workflows for Curation and Analysis of Microstructure-Aware Materials Data: Application to Aging of U-Nb Alloys

Questions about ProgramMaster? Contact programming@programmaster.org