Automated Copper Alloy Grain Size Evaluation Using a Deep-learning CNN

Moog Inc. has automated the evaluation of copper (Cu) alloy grain size using a deep-learning convolutional neural network (CNN). The proof-of-concept automated image acquisition and batch-wise image processing offers the potential for significantly reduced labor, improved accuracy of grain evaluation, and decreased overall turnaround times for approving Cu alloy bar stock for use in flight critical aircraft hardware. A classification accuracy of 91.1% on individual sub-images of the Cu alloy coupons was achieved. Process development included minimizing the variation in acquired image color, brightness, and resolution to create a dataset with 12300 sub-images, and then optimizing the CNN hyperparameters on this dataset using statistical design of experiments (DoE). Over the development of the automated Cu alloy grain size evaluation, a degree of "explainability" in the artificial intelligence (XAI) output was realized, based on the decomposition of the large raw images into many smaller dataset sub-images, through the ability to explain the CNN ensemble image output via inspection of the classification results from the individual smaller sub-images.

[1]  François Chollet,et al.  Deep Learning with Python , 2017 .

[2]  Limei Song,et al.  Detection of Micro-Defects on Metal Screw Surfaces Based on Deep Convolutional Neural Networks , 2018, Sensors.

[3]  Luis Perez,et al.  The Effectiveness of Data Augmentation in Image Classification using Deep Learning , 2017, ArXiv.

[4]  Sebastian Ruder,et al.  An overview of gradient descent optimization algorithms , 2016, Vestnik komp'iuternykh i informatsionnykh tekhnologii.

[5]  Sepp Hochreiter,et al.  Self-Normalizing Neural Networks , 2017, NIPS.

[6]  Christopher J. Nachtsheim,et al.  Definitive Screening Designs with Added Two-Level Categorical Factors* , 2013 .

[7]  George E. P. Box,et al.  Evolutionary Operation: A Statistical Method for Process Improvement , 1969 .

[8]  Jorge Nocedal,et al.  On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima , 2016, ICLR.

[9]  Prakhar Jaiswal,et al.  Convolutional neural network-based inspection of metal additive manufacturing parts , 2019, Rapid prototyping journal.

[10]  L. F. Abbott,et al.  Random Walk Initialization for Training Very Deep Feedforward Networks , 2014, 1412.6558.

[11]  Carlo Luschi,et al.  Revisiting Small Batch Training for Deep Neural Networks , 2018, ArXiv.

[12]  Timothy Dozat,et al.  Incorporating Nesterov Momentum into Adam , 2016 .

[13]  J. S. Dahmann,et al.  Future directions of C3 research at DARPA (Defense Advanced Research Projects Agency) , 1987 .

[14]  Riska Asriana Sutrisnowati,et al.  GEAR CLASSIFICATION FOR DEFECT DETECTION IN VISION INSPECTION SYSTEM USING DEEP CONVOLUTIONAL NEURAL NETWORKS , 2018 .

[15]  Petr Kuritcyn,et al.  Using artificial intelligence strategies for process-related automated inspection in the production environment , 2016 .

[16]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[17]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[18]  Two-Phase Grain Structures Standard Test Methods for Determining Average Grain Size Using Semiautomatic and Automatic Image Analysis 1 , 2004 .

[19]  Nikhil Ketkar,et al.  Deep Learning with Python , 2017 .

[20]  이선영,et al.  Histogram Equalization , 2012, Encyclopedia of Biometrics.

[21]  Christopher J. Nachtsheim,et al.  A Class of Three-Level Designs for Definitive Screening in the Presence of Second-Order Effects , 2011 .

[22]  Jay Lee,et al.  Quality analysis in metal additive manufacturing with deep learning , 2020, Journal of Intelligent Manufacturing.