Image Inspection of Railcar Structural Components: An approach through Deep Learning and Discrete Fourier Transform

• Rafael L. Rocha Universidade Federal do Pará / Instituto SENAI de Inovação em Tecnologias Minerais
• Cleison D. Silva Universidade Federal do Pará
• Ana Claudia S. Gomes Instituto SENAI de Inovação em Tecnologias Minerais
• Bruno V. Ferreira Instituto SENAI de Inovação em Tecnologias Minerais
• Eduardo C. Carvalho Instituto SENAI de Inovação em Tecnologias Minerais
• Ana Carolina Q. Siravenha Instituto Tecnológico Vale
• Schubert R. Carvalho Instituto Tecnológico Vale

DOI: https://doi.org/10.5753/kdmile.2019.8786

Resumo

Railcar components inspection is one of the most critical tasks in railway maintenance. The use of image processing, coupled with machine learning, has emerged as a solution for replacing current standard methodologies. The spectral analysis gives the frequency representation of a signal and has been largely used in signal processing tasks. In this sense, this work proposes the evaluation of the use of the Discrete Fourier Transform (DFT) in addition to the spatial representation image of railcar components for an automatic detector of defective parts performed by Convolutional Neural Network (CNN) classi cation. The results are given in measures of accuracy, precision, recall, and F1-score metrics in addition to the accuracy boxplot, and showed that the use of the DFT increase in 1.04% the CNN classification accuracy.

Palavras-chave: railcar inspection, convolutional neural network, discrete Fourier transform, image classi cation

Referências

Bishop, C. M. Pattern recognition and machine learning. Springer, 2006.

Gibert, X., Patel, V. M., and Chellappa, R. Deep multitask learning for railway track inspection. IEEE Transactions on Intelligent Transportation Systems 18 (1): 153164, 2017.

Gonzalez, R. C. and Woods, R. E. Digital image processing. Prentice Hall, Upper Saddle River, N.J., 2006.

Goodfellow, I., Bengio, Y., and Courville, A. Deep Learning. MIT Press, 2016. http://www.deeplearningbook.org.

Hart, J., Resendiz, E., Freid, B., Sawadisavi, S., Barkan, C., and Ahuja, N. Machine vision using multi-spectral imaging for undercarriage inspection of railroad equipment. In Proceedings of the 8th World Congress on Railway Research, Seoul, Korea, 2008.

Haykin, S. Neural networks and learning machines. Pearson Prentice Hall, Upper Saddle River, N.J., 2009. IWnicki, S. Handbook of Railway Vehicle Dynamics. CRC Press, 2006.

Liu, L., Zhou, F., and He, Y. Automated visual inspection system for bogie block key under complex freight train environment. IEEE Transactions on Instrumentation and Measurement 65 (1): 214, 2016.

Macucci, M., Di Pascoli, S., Marconcini, P., and Tellini, B. Derailment detection and data collection in freight trains, based on a wireless sensor network. IEEE Transactions on Instrumentation and Measurement 65 (9):19771987, 2016.

Park, B., Chen, Y., Nguyen, M., and Hwang, H. Characterizing multispectral images of tumorous, bruised, skin-torn, and wholesome poultry carcasses. Transactions of the ASAE 39 (5): 19331941, 1996.

Park, J.-K., Kwon, B.-K., Park, J.-H., and Kang, D.-J. Machine learning-based imaging system for surface defect inspection. International Journal of Precision Engineering and Manufacturing-Green Technology 3 (3): 303310, Jul, 2016.

Ravikumar, S., Ramachandran, K. I., and Sugumaran, V. Machine Learning Approach for Automated Visual Inspection of Machine Components. Expert Syst. Appl. 38 (4): 32603266, 2011.