Research on Self-organized CNN Modeling to Identify Metallographic Structure of Heat-resistant Steel
ZHANG Yongzhi1, LI Xuying1, XIN Quanzhong2, KONG Xiangming2, WANG Yongliang2
1 College of Mechanical and Electrical Engineering, Inner Mongolia Agricultural University, Hohhot 010018, China 2 Electric Power Engineering and Technology Institute, Inner Mongolia Energy Power Investment Group Co., Ltd., Hohhot 010090, China
Abstract: Heat-resistant steel has complex metallographic structure, hence it is difficult to extract features using traditional image processing methods to accurately achieve auto recognition. Manual recognition is susceptible to subjective factors and thus resulting in poor accuracy and repeatability. Convolutional neural networks (CNN) can extract complex features from raw images, only with the problem of difficulty in model training and selection and optimization of topological hyperparameters. In this work, a Hyperband algorithm, which calculates resource allocation based on hyperparameter sets, was used to optimize hyperparameter in CNN model and finally the self-organizing modelling was achieved. Relevant problems were solved such as low efficiency and large consumption in computing resources due to grid searching, random searching and Bayesian optimization as well as optimization instability. Based on above-mentioned Hyperband algorithm optimization, a 33 layers' CNN model was constructed, trained, simulated and its recognition performance evaluated with confusion matrix analysis. Results showed that, in recognition of metallographic structure for heat-resistant steel, an average accuracy of 94.2%, an average precision of 94.1%, an average sensitivity of 94.2 and an average specificity of 98.1% were obtained after using the established model. Our model demonstrated a good generalization ability and could be used to recognize metallagraphic structure accurately. This approach provided a new method in intelligent recognition of complex metallagraphic structures.
1 Sosa J M, Huber D E, Welk B, et al. Integrating Materials & Manufacturing Innovation, 2014, 3(1), 1. 2 Kotas P, Praks P, Valek L, et al. Advances in Electrical & Electronic Engineering, 2012, 10(1), 50. 3 Domínguez R G, Gonzalez S J A, Rosado C J, et al. Metals and Materials International, 2019, 7, 1. 4 Ontman A Y M, Shiflet G J. JOM: the Journal of the Minerals, Metals & Materials Society, 2011, 63(7), 44. 5 Magdeev R G, Tashlinskiy A G. In: 3rd International Conference Information Technology and Nanotechnology. Samara, 2017, pp. 158. 6 Iacoviello F, De S A, Iacoviello D, et al. Metallurgia Italiana, 2007, 99(6), 15. 7 Lecun Y, Bengio Y, Hinton G E. Nature, 2015, 521(7553), 436. 8 Decost B L, Lei B, Francis T, et al. Microsc Microanal, 2019, 25(1), 21. 9 Chowdhury A, Kautz E, Yener B, et al. Computational Materials Science, 2016, 123, 176. 10 Decost B L, Holm E A. Computational Materials Science, 2015, 110, 126. 11 Naik D L, Sajid H U, Kiran R. Metals-Open Access Metallurgy Journal, 2019, 9(5), 546. 12 Kusche C, Reclik T, Freund M, et al. PLos One, 2019, 14(5), 1. 13 Wang L Y, Tong Z N, Liu Y W, et al. In: The 16th China Conference on Stereology and Image Analysis. Haikou, China, 2019, pp. 330 (in Chinese). 王乐耘, 童周诺, 刘言伟,等. 第十六届中国体视学与图像分析学术会议.海口, 2019, pp. 330. 14 Hu F K. Microstructure recognition of high-strength steel materials based on texture recognition and deep convolutional neural network.Master's Thesis, Huazhong University of Science and Technology, China, 2019 (in Chinese). 胡方康. 基于纹理识别和深度卷积神经网络的高强钢材料显微组织识别. 硕士学位论文, 华中科技大学, 2019. 15 Dimiduk D M, Holm E A, Niezgoda S R. Integrating Materials and Manufacturing Innovation, 2018, 7(3), 157. 16 Decost B L, Francis T, Holm E A. Acta Materialia, 2017, 133, 30. 17 Adachi Y, Taguchi M, Hirokawa S. Tetsu to Hagane, 2016, 102(12), 722. 18 Srivastava R K, Greff K, Schmidhuber J. In: NIPS'15: Proceedings of the 28th International Conference on Neural Information Processing Systems. Montreal, Canada, 2015, pp. 2377. 19 Ioffe S, Szegedy C. In: ICML'15: Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille, 2015, pp. 448. 20 Domhan T, Springenberg J T, Hutter F. In: IJCAI'15: Proceedings of the 24th International Conference on Artificial Intelligence. Buenos Aires, 2015, pp. 3460. 21 Hesterman J Y, Caucci L, Kupinski M A, et al. IEEE Transactions on Nuclear Science, 2010, 57(3), 1077. 22 Solis F J, Wets R J B. Mathematics of Operations Research, 1981, 6(1), 19. 23 Spall J C. Introduction to stochastic search and optimization: estimation, simulation, and control, John Wiley & Sons, Inc Press, US, 2003. 24 Bergstra J, Bengio Y. Journal of Machine Learning Research, 2012, 13, 281. 25 Jones D R, Schonlau M, Welch W J. Journal of Global Optimization, 1998, 13(4), 455. 26 Snoek J, Larochelle H, Adams R P. Advances in Neural Information Processing Systems, 2012, 25, 2960. 27 Snoek J, Rippel O, Swersky K, et al. Statistics, 2015, 57(3), 1861. 28 Springenberg J T, Klein A, Falkner S, et al. In: NIPS'16: Proceedings of the 30th International Conference on Neural Information Processing Systems.Barcelona, Spain, 2016, pp. 4141. 29 Kandasamy K, Schneider J, Poczos B. In: ICML'15: Proceedings of the 32nd International Conference on International Conference on Machine Learning. Lille, France, 2015, pp. 295. 30 Wang Z Y, Zoghi M, Hutter F, et al. In: IJCAI'13: Proceedings of the Twenty-Third International Joint Conference on Artificial Intelligence. Beijing, China, 2013, pp. 1778. 31 Li L S, Jamieson K, Desalvo G, et al. Journal of Machine Learning Research, 2018, 18, 1. 32 Jamieson K, Talwalkar A. In: the 19th International Conference on Artificial Intelligence and Statistics. Guilin, China, 2016, pp. 240. 33 Simonyan K, Zisserman A. arXiv Preprint arXiv, DOI:org/10.48550/arXiv.1409.1556 34 He K M, Zhang X Y, Ren S Q, et al. In: ICCV'15: Proceedings of the 2015 IEEE International Conference on Computer Vision. Santiago, Chile, 2015, pp. 1026. 35 Foody G M. Remote Sensing of Environment, 2002, 80(1), 185.