Search Menu
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information
Search Scholarly Publications

Title: A study of real-world micrograph data quality and machine learning model robustness

Journal Article · · npj Computational Materials
ORCiD logo; ; ORCiD logo; ; ; ORCiD logo

Abstract Machine-learning (ML) techniques hold the potential of enabling efficient quantitative micrograph analysis, but the robustness of ML models with respect to real-world micrograph quality variations has not been carefully evaluated. We collected thousands of scanning electron microscopy (SEM) micrographs for molecular solid materials, in which image pixel intensities vary due to both the microstructure content and microscope instrument conditions. We then built ML models to predict the ultimate compressive strength (UCS) of consolidated molecular solids, by encoding micrographs with different image feature descriptors and training a random forest regressor, and by training an end-to-end deep-learning (DL) model. Results show that instrument-induced pixel intensity signals can affect ML model predictions in a consistently negative way. As a remedy, we explored intensity normalization techniques. It is seen that intensity normalization helps to improve micrograph data quality and ML model robustness, but microscope-induced intensity variations can be difficult to eliminate.

Sponsoring Organization:
USDOE
OSTI ID:
1823683
Journal Information:
npj Computational Materials, Journal Name: npj Computational Materials Journal Issue: 1 Vol. 7; ISSN 2057-3960
Publisher:
Nature Publishing GroupCopyright Statement
Country of Publication:
United Kingdom
Language:
English

References (49)

A comparison of conventional Everhart-Thornley style and in-lens secondary electron detectors-a further variable in scanning electron microscopy journal May 2011
Electron Backscatter Diffraction in Materials Science book January 2009
KAZE Features book January 2012
Computer Vision and Machine Learning for Autonomous Characterization of AM Powder Feedstocks journal December 2016
Machine Learning for Automated Quality Evaluation in Pharmaceutical Manufacturing of Emulsions journal April 2019
Microstructure Cluster Analysis with Transfer Learning and Unsupervised Learning journal August 2018
Adaptive histogram equalization and its variations journal September 1987
Exploring the microstructure manifold: Image texture representations applied to ultrahigh carbon steel microstructures journal July 2017
Microstructure recognition using convolutional neural networks for prediction of ionic conductivity in ceramics journal December 2017
A computer vision approach for automated analysis and classification of microstructural image data journal December 2015
Image driven machine learning methods for microstructure recognition journal October 2016
Improving direct physical properties prediction of heterogeneous materials from imaging data via convolutional neural network and a morphology-aware generative model journal July 2018
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI journal June 2020
Predicting compressive strength of consolidated molecular solids using computer vision and deep learning journal May 2020
Building data-driven models with microstructural images: Generalization and interpretability journal December 2017
A methodology of steel microstructure recognition using SEM images by machine learning based on textural analysis journal December 2020
Artificial intelligence for the prediction of tensile properties by using microstructural parameters in high strength steels journal June 2020
Computational microstructure characterization and reconstruction: Review of the state-of-the-art techniques journal June 2018
Random Forests journal January 2001
Distinctive Image Features from Scale-Invariant Keypoints journal November 2004
Realization of the Contrast Limited Adaptive Histogram Equalization (CLAHE) for Real-Time Image Enhancement journal August 2004
Machine learning in materials informatics: recent applications and prospects journal December 2017
Advanced Steel Microstructural Classification by Deep Learning Methods journal February 2018
Image-driven discriminative and generative machine learning algorithms for establishing microstructure–processing relationships journal October 2020
Six decades of the Hall–Petch effect – a survey of grain-size strengthening studies on pure metals journal June 2016
Automatic steel labeling on certain microstructural constituents with image processing and machine learning tools journal June 2019
Toward Fast Calibration of Global Drift in Scanning Electron Microscopes with Respect to Time and Magnification journal January 2012
ImageNet: A large-scale hierarchical image database
  • Deng, Jia; Dong, Wei; Socher, Richard
  • 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2009 IEEE Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.2009.5206848
conference June 2009
Aggregating local descriptors into a compact image representation
  • Jegou, Herve; Douze, Matthijs; Schmid, Cordelia
  • 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.2010.5540039
conference June 2010
All About VLAD conference June 2013
Analyzing Classifiers: Fisher Vectors and Deep Neural Networks conference June 2016
Deep Residual Learning for Image Recognition conference June 2016
Densely Connected Convolutional Networks conference July 2017
An Analysis of Scale Invariance in Object Detection - SNIP conference June 2018
CNN Features Off-the-Shelf: An Astounding Baseline for Recognition conference June 2014
An HOG-LBP human detector with partial occlusion handling conference September 2009
ORB: An efficient alternative to SIFT or SURF conference November 2011
Steel defect classification with Max-Pooling Convolutional Neural Networks
  • Masci, Jonathan; Meier, Ueli; Ciresan, Dan
  • 2012 International Joint Conference on Neural Networks (IJCNN 2012 - Brisbane), The 2012 International Joint Conference on Neural Networks (IJCNN) https://doi.org/10.1109/IJCNN.2012.6252468
conference June 2012
Perceptual Color Image Coding With JPEG2000 journal February 2010
Least squares quantization in PCM journal March 1982
SIFT Meets CNN: A Decade Survey of Instance Retrieval journal May 2018
Anomaly detection: A survey journal July 2009
ImageNet classification with deep convolutional neural networks journal May 2017
The Mythos of Model Interpretability: In machine learning, the concept of interpretability is both important and slippery. journal June 2018
Prototype selection for interpretable classification journal December 2011
Survey of Contrast Enhancement Techniques based on Histogram Equalization journal January 2011
On The Effect Of Image Brightness And Contrast Nonuniformity On Statistical Texture Parameters journal September 2015
UMAP: Uniform Manifold Approximation and Projection journal September 2018
Wide Residual Networks conference January 2016

Similar Records

Related Subjects