METAL PLAKA YÜZEY KUSURLARININ TESPİTİNDE DERİN ÖĞRENME UYGULAMASI VE ANALİZİ

Yıl 2024, Cilt: 5 Sayı: 2, 263 - 285, 20.12.2024

Can Tuncer , Cemil Közkurt , Serhat Kılıçarslan

https://doi.org/10.55546/jmm.1512549

Öz

Endüstriyel imalat proseslerinde, demir çelik ana sanayi üreticilerinden talaşlı ve talaşsız yöntemlerle işlenmek üzere temin edilen metal levhaların yüzeylerindeki kusurların tespiti, ilgili levhanın güvenlik ve bakım maliyeti gibi değerlerinin tahmin edilmesinde önemli bir yer tutmaktadır. Gelişen teknolojiyle bilgisayarlı görü ve derin öğrenme uygulamalarının endüstride kendine yer bulması ile metal plaka yüzey kusurlarının ileri teknolojik düzeyde daha hızlı ve etkin bir şekilde daha düşük hata oranıyla tespit edilmesi ve sınıflandırılması mümkün hale gelmiştir. Bu çalışma kapsamında, metal plaka yüzey kusurlarını tespit etmek için NEU Metal Yüzey Kusurları Veri Seti kullanılarak Python ortamında TensorFlow kütüphanesi kullanılarak bir derin öğrenme modeli oluşturulmuştur. Daha sonra endüstriyel uygulama olarak bu modeli gerçek koşullar altında test etmek amacıyla Nvidia Jetson Nano ve USB Kamera kullanılarak bir cihaz prototipi geliştirilmiştir.

Anahtar Kelimeler

Metal Plaka, Yüzey Kusuru, Derin Öğrenme, Bilgisayarlı görü, Yapay Zeka, Makine Öğrenmesi

Etik Beyan

Etik kurallara uyum gerektirecek herhangi bir süreç, canlı varlıklar, organizasyonlar, kurumlar üzerinde herhangi bir çalışma gerçekleştirilmemiştir.

Destekleyen Kurum

Bandırma Onyedi Eylül Üniversitesi Bilimsel Araştırma Projeleri Birimi

Proje Numarası

BAP 22-1010-002

Teşekkür

Yürütmekte olduğumuz BAP 22-1010-002 numaralı projemiz boyunca verdikleri destek için Bandırma Onyedi Eylül Üniversitesi Bilimsel Araştırma Projeleri Birimi’ne teşekkürlerimizi sunarız.

Deep Learning Application and Analysis In Detection of Metal Plate Surface Defects

Öz

In industrial manufacturing processes, detection of defects on the surfaces of metal plates supplied from iron and steel main industry manufacturers to be processed by machining and non-machining methods has an important place in estimating the values of the relevant plate such as safety and maintenance cost. With the developing technology and computer vision and deep learning applications finding a place in the industry, it has become possible to detect and classify metal plate surface defects more quickly and effectively with a lower error rate at an advanced technological level. Within the scope of this study, a deep learning model was created by using the TensorFlow library in the Python environment with using NEU Metal Surface Defects Dataset to detect metal plate surface defects. Then as an industrial application, a device prototype developed using Nvidia Jetson Nano and USB Camera, in order to test this model under real conditions.

Anahtar Kelimeler

Metal Plate, Surface Defect, Deep Learning, Computer Vision, Artificial Intelligence, Machine Learning

Etik Beyan

No work has been carried out on any process, living beings, organizations or institutions that would require compliance with ethical rules.

Destekleyen Kurum

Bandırma Onyedi Eylül University Scientific Research Projects Unit

Proje Numarası

BAP 22-1010-002

Teşekkür

We would like to thank Bandırma Onyedi Eylül University Scientific Research Projects Unit for their support throughout our project numbered BAP 22-1010-002.

Kaynakça

• Agarwal M., Gupta S., Biswas K. K., A new Conv2D model with modified ReLU activation function for identification of disease type and severity in cucumber plant. Sustainable Computing: Informatics and Systems 30, 100473, 2021.
• Baldi P., Sadowski P. J., Understanding Dropout. Advances in Neural Information Processing Systems 26, 2013. Barz B., Denzler J., Deep Learning on Small Datasets without Pre-Training using Cosine Loss, In Proceedings of the IEEE/CVF winter conference on applications of computer vision, 2020, pp: 1371-1380.
• Bbouzidi S., Hcini G., Jdey I., Drira F., Convolutional Neural Networks and Vision Transformers for Fashion MNIST Classification: A Literature Review. arXiv preprint arXiv:2406.03478, 2024.
• Bock S., Weiß M., A Proof of Local Convergence for the Adam Optimizer. 2019 International Joint Conference on Neural Networks (IJCNN), July, 2019, pp: 1-8.
• Dung L., Mizukawa M., A Pattern Recognition Neural Network Using Many Sets of Weights and Biases. 2007 International Symposium on Computational Intelligence in Robotics and Automation, 2007, pp: 285–290. Glassmacher S., https://galaxyinferno.com/epochs-iterations-and-batch-size-deep-learning-basics-explained/, 2022, (24 October 2022).
• Gulli A., Pal S., Deep Learning with Keras. Packt Publishing Ltd., 2017.
• Haji S. H., Abdulazeez A. M., Comparison of Optimization Techniques Based on Gradient Descent Algorithm: A Review. PalArch’s Journal of Archaeology of Egypt / Egyptology 18(4), 2715-2743, 2021.
• Helms M., Ault S. V., Mao G., Wang J., An Overview of Google Brain and Its Applications. Proceedings of the 2018 International Conference on Big Data and Education, March, 2018, pp: 72-75.
• Jin J., Dundar A., Culurciello E., Flattened Convolutional Neural Networks for Feedforward Acceleration, arXiv preprint arXiv:1412.5474,2015.
• Kılıçarslan S., Adem K., Çelik M., An overview of the activation functions used in deep learning algorithms. Journal of New Results in Science, 10(3), 2021.
• LeCun Y., Bengio Y., Hinton G., Deep learning. Nature 521(7553), 436-444, 2015.
• Lv X., Duan F., Jiang J., Fu X., Gan L., Deep Metallic Surface Defect Detection: The New Benchmark and Detection Network. Sensors 20(6) 1562, 2020.
• Maharana K., Mondal S., Nemade B., A review: Data pre-processing and data augmentation techniques. Global Transitions Proceedings 3(1), 91-99, 2022.
• Manaswi N. K., Understanding and Working with Keras. In N. K. Manaswi (Ed.), Deep Learning with Applications Using Python: Chatbots and Face, Object, and Speech Recognition With TensorFlow and Keras, Apress, pp:1–43, 2018.
• Mastromichalakis S. Parametric Leaky Tanh: A New Hybrid Activation Function for Deep Learning arXiv preprint arXiv:2310.07720, 2023.
• Maxwell A. E., Warner T. A., Guillén L. A., Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies—Part 2: Recommendations and Best Practices. Remote Sensing 13(13), 2591, 2021.
• Mo X., Tao K., Wang Q., Wang G., An Efficient Approach for Polyps Detection in Endoscopic Videos Based on Faster R-CNN. 2018 24th International Conference on Pattern Recognition (ICPR), August, 2018, pp: 3929-3934.
• Montesinos López O. A., Montesinos López A., Crossa J., Overfitting, model tuning, and evaluation of prediction performance. In Multivariate statistical machine learning methods for genomic prediction. Cham: Springer International Publishing, pp: 109-139, 2022.
• Moor J., The Dartmouth College Artificial Intelligence Conference: The Next Fifty Years. AI Magazine 27(4), 87-87, 2006.
• Oranen L., Utilizing deep learning on embedded devices, 2021.
• Rasamoelina A. D., Adjailia F., Sinčák P., A Review of Activation Function for Artificial Neural Network. 2020 IEEE 18th World Symposium on Applied Machine Intelligence and Informatics (SAMI), 281–286, 2020.
• Raschka S., Mirjalili V., Python Machine Learning: Machine Learning and Deep Learning with Python, scikit-learn, and TensorFlow 2. Packt Publishing Ltd., 2019.
• Rice L., Wong E., Kolter Z., Overfitting in adversarially robust deep learning. Proceedings of the 37th International Conference on Machine Learning, November, 2020, pp: 8093-8104.
• Sahu M., Dash R., A Survey on Deep Learning: Convolution Neural Network (CNN). In D. Mishra, R. Buyya, P. Mohapatra, & S. Patnaik (Eds.), Intelligent and Cloud Computing, Springer, pp: 317-325, 2021.
• Shinde P. P., Shah S., A Review of Machine Learning and Deep Learning Applications. 2018 Fourth International Conference on Computing Communication Control and Automation (ICCUBEA), August, 2018, pp: 1-6.
• Smith S. L., Kindermans P. J., Ying C., Le Q. V., Don’t Decay the Learning Rate, Increase the Batch Size arXiv preprint arXiv:1711.00489v2, 2018.
• Sobhana M., Hindu K., Girishma N., Bhavani P. S., Rajeswari S., A Comparitive Evaluation Of Custom CNN, Sequential CNN & Dense-Net Models to forecast Dementia. 2023 Second International Conference on Augmented Intelligence and Sustainable Systems (ICAISS), 2023 pp: 287-293,2023.
• Song K., Yunhui Y., NEU surface defect database. Northeastern University, 2019.
• Swasthik, “Understanding Entropy and Losses and all those confusing names of losses,”, 2020, https://medium.com/@swasthik0304/understanding-entropy-and-losses-and-all-those-confusing-names-of-losses-d7444711cf3c>, (21 April 2020).
• Turban E., Watkins P. R., Integrating Expert Systems and Decision Support Systems. MIS Quarterly, 10(2), 121-136, 1986.
• Vansh, “Predicting House Prices Using Keras Functional API” 2022, <https://www.analyticsvidhya.com/blog/2022/04/predicting-house-prices-using-keras-functional-api/>, (9 May 2022).
• Wang M., Lu S., Zhu D., Lin J., Wang Z., A High-Speed and Low-Complexity Architecture for Softmax Function in Deep Learning. 2018 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS), October, 2018, pp: 223-226.
• Wang Q., Ma Y., Zhao K., Tian Y., A Comprehensive Survey of Loss Functions in Machine Learning. Annals of Data Science 9(2), 187-212, 2022.
• Wang S., Xia X., Ye L., Yang B., Automatic detection and classification of steel surface defect using deep convolutional neural networks. Metals 11 (3), 1-23, 2021.
• Warden, Pete, “Why GEMM is at the heart of deep learning” 2015, <https://petewarden.com/2015/04/20/why-gemm-is-at-the-heart-of-deep-learning/>, (20 April 2015)
• Yun J. P., Shin W. C., Koo G., Kim M. S., Lee C., Lee S. J., Automated defect inspection system for metal surfaces based on deep learning and data augmentation. Journal of Manufacturing Systems 55, 317-324, 2020.
• Zhao W., Chen F., Huang H., Li D., Cheng W., A new steel defect detection algorithm based on deep learning. Computational Intelligence and Neuroscience 2021(1), 5592878, 2021.