AHŞAP HAM MADDELERDE YÜZEY HATASINI BELİRLEMEK İÇİN GÖRÜNTÜ İŞLEME TABANLI KALİTE KONTROL SİSTEMİ

Öz

Günümüzde ahşap ham madde malzemeleri birçok endüstride kullanılmaktadır. Ahşap ham madde üzerinde gözden kaçan kusurlar, son ürünü görsel açıdan ve dayanıklılık bakımından etkileyerek satışını engelleyebilir. Kusurlu ham maddeler üzerinde elle veya görsel kontrol zor ve yanıltıcı olabilir. Sürekli gelişen dijital teknoloji ve akıllı sistemler sayesinde, kalite kontrol için otomasyon sistemleri geliştirilmektedir. Böylece üretimin erken aşamalarında kusurlar tespit edilebilmektedir. Son ürünün kusurlu veya hatalı olması engellenebildiğinde iş gücü, malzeme ve zaman kayıpları önlenebilecek, maliyetler düşürülebilecektir. Bu çalışmada, özel bir kamera sistemi ile elde edilen görüntüler, görüntü işleme temelli Derin Öğrenme (DÖ) yöntemlerinde kullanılarak ahşap ham maddenin kusurlu olup olmadığı ayrımı yapılmaktadır. Kusurların tespitinde kullanılacak model ayrım odaklı bir yöntem olan Evrişimsel Sinir Ağı (ESA) ile geliştirilmiş olup tercih edilen bir yöntemdir. Çalışmada, ShuffleNet, AlexNet, GoogleNet gibi ESA mimarileri denenmiş ve en uygun mimari belirlenmiştir. Çalışmalar sonucunda, ESA mimarileri arasında kusurlu ve kusursuz ahşap ürünlerini belirlemek için kullanılan başarılı yöntemler olan MobileNet, DenseNet ve Inception mimarilerinin, kalite kontrol sistemleri için daha yüksek doğruluk oranları ile sonuçlandığı görülmüştür. En iyi sonuç ise, Inception-V3 mimarisi ile elde edilmiştir ve test doğruluğu %88,41 olarak belirlenmiştir.

Anahtar Kelimeler

• Derin Öğrenme
• Görüntü İşleme
• Ahşap Kusur Tespiti
• Evrişimsel Sinir Ağları
• Sınıflandırma

IMAGE PROCESSING-BASED QUALITY CONTROL SYSTEM TO DETERMINE THE SURFACE DEFECT IN WOODEN RAW MATERIALS

Öz

In modern industries, wood raw materials serve a multitude of purposes across various sectors. Undetected flaws within these materials can significantly impact products, affecting both their visual appeal and durability, leading to compromised marketability. Manual and visual inspection of flawed raw materials is a complex process prone to errors in judgment. The rapid evolution of digital technologies has spurred the creation of automated systems designed for precise quality assessments. This progress enables the early identification of defects during production, thereby preventing the occurrence of defective or substandard end-products. Consequently, this approach reduces labor, material, and time wastage, effectively cutting down associated costs.The present study focuses on distinguishing between defective and intact raw materials using images obtained through a specialized camera system. Deep learning techniques in image processing are employed for this purpose, with a particular emphasis on the Convolutional Neural Network (CNN), a classification method chosen for defect detection. A range of CNN architectures, including ShuffleNet, AlexNet, and GoogleNet, were tested, and the most effective one was identified. The results of these experiments demonstrate that within the realm of CNNs, architectures like MobileNet, DenseNet, and Inception have proven highly successful, leading to improved accuracy in quality control systems. Notably, the Inception-V3 architecture emerges as the top performer, achieving a test accuracy of 88.41%.

Anahtar Kelimeler

• Deep Learning
• Image Processing
• Wood Defect Detection
• Convolutional Neural Network
• Classification

Kaynakça

1. Aparecido De França, C., & Gonzaga, A. (2005). Classification of Wood Plates by Neural Networks and Fuzzy Logic Image and Video Processing View project Detection of Architectural Distortion in Mammograms View project. Computer Science. https://www.researchgate.net/publication/266290506
2. Cavalin, P., Oliveira, L. S., Koerich, A. L., & Britto, A. S. (2006). Wood defect detection using grayscale images and an optimized feature set. IECON Proceedings (Industrial Electronics Conference), 3408-3412. https://doi.org/10.1109/IECON.2006.347618
3. Deng, J., Dong, W., Socher, R., Li, L.-J., Kai Li, & Li Fei-Fei. (2009). ImageNet: A large-scale hierarchical image database. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 248-255. https://doi.org/10.1109/CVPR.2009.5206848
4. Ding, F., Zhuang, Z., Liu, Y., Jiang, D., Yan, X., & Wang, Z. (2020). Detecting defects on solid wood panels based on an improved SSD algorithm. Sensors (Switzerland), 20(18), 1-17. https://doi.org/10.3390/S20185315
5. Faura, Á. G., Štepec, D., Cankar, M., & Humar, M. (2021). Application of unsupervised anomaly detection techniques to moisture content data fromwood constructions. Forests, 12(2), 1-19. https://doi.org/10.3390/F12020194
6. Fawcett, T. (2004). ROC Graphs: Notes and Practical Considerations for Researchers. Pattern Recognition Letters, 31(8), 1-38. https://www.researchgate.net/publication/284043217_ROC_Graphs_Notes_and_Practical_Considerations_for_Researchers
7. Fırıldak, K., & Talu, M. F. (2019). Evrişimsel Sinir Ağlarında Kullanılan Transfer Öğrenme Yaklaşımlarının İncelenmesi. Anatolian Journal of Computer Science - Anatolian Science, 4(2), 88-95. https://dergipark.org.tr/tr/pub/bbd/issue/49546/527863
8. Fukushima, K. (1980). Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics 1980 36:4, 36(4), 193-202. https://doi.org/10.1007/BF00344251

9. Gao, M., Chen, J., Mu, H., & Qi, D. (2021). A Transfer Residual Neural Network Based on ResNet-34 for Detection of Wood Knot Defects. Forests 2021, Vol. 12, Page 212, 12(2), 227. https://doi.org/10.3390/F12020212
10. He, T., Liu, Y., Xu, C., Zhou, X., Hu, Z., & Fan, J. (2019). A fully convolutional neural network for wood defect location and identification. IEEE Access, 7, 123453-123462. https://doi.org/10.1109/ACCESS.2019.2937461
11. Hinton, G. E., Osindero, S., & Teh, Y. W. (2006). A Fast Learning Algorithm for Deep Belief Nets. Neural Computation, 18(7), 1527-1554. https://doi.org/10.1162/NECO.2006.18.7.1527
12. Huang, G., Liu, Z., van der Maaten, L., & Weinberger, K. Q. (2017). Densely connected convolutional networks. Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, 2017-January, 2261-2269. https://doi.org/10.1109/CVPR.2017.243
13. İnik, Ö., & Ülker, E. (2017). Derin Öğrenme ve Görüntü Analizinde Kullanılan Derin Öğrenme Modelleri. Gaziosmanpaşa Bilimsel Araştırma Dergisi, 6(3), 85-104. https://dergipark.org.tr/tr/pub/gbad/issue/31228/330663
14. Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., & Keutzer, K. (2016, Şubat 24). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size. ICLR 2017. https://doi.org/10.48550/arxiv.1602.07360
15. Kamal, K., Qayyum, R., Mathavan, S., & Zafar, T. (2017). Wood defects classification using laws texture energy measures and supervised learning approach. Advanced Engineering Informatics, 34, 125-135. https://doi.org/10.1016/J.AEI.2017.09.007
16. Karaca, B. K., Guney, S., Dengiz, B., & Agildere, M. (2021). Comparative Study for Tuberculosis Detection by Using Deep Learning. 2021 44th International Conference on Telecommunications and Signal Processing, TSP 2021, 88-91. https://doi.org/10.1109/TSP52935.2021.9522634
17. Kesici, B., & Yildiz, M. S. (2016). Kalite Kontrol Faaliyetlerinde Yapay Zekâ Kullanımı ve Bir Otomotiv Yan Sanayisinde Uygulanması. Yalova Sosyal Bilimler Dergisi, 6(12), 307-323. https://dergipark.org.tr/tr/pub/yalovasosbil/issue/27392/289024
18. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84-90. https://doi.org/10.1145/3065386
19. Mahram, A., Shayesteh, M. G., & Jafarpour, S. (2012). Classification of wood surface defects with hybrid usage of statistical and textural features. 2012 35th International Conference on Telecommunications and Signal Processing, TSP 2012 - Proceedings, 749-752. https://doi.org/10.1109/TSP.2012.6256397
20. Masci, J., Meier, U., Ciresan, D., Schmidhuber, J., & Fricout, G. (2012). Steel defect classification with Max-Pooling Convolutional Neural Networks. Proceedings of the International Joint Conference on Neural Networks. https://doi.org/10.1109/IJCNN.2012.6252468
21. McCulloch, W. S., & Pitts, W. (1943). A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics 1943 5:4, 5(4), 115-133. https://doi.org/10.1007/BF02478259
22. Mohan, S., & Venkatachalapathy, K. (2012). Wood Knot Classification using Bagging. International Journal of Computer Applications, 51(18), 50-53. https://doi.org/10.5120/8146-1937
23. Perez-Cortes, J. C., Perez, A. J., Saez-Barona, S., Guardiola, J. L., & Salvador, I. (2018). A System for In-Line 3D Inspection without Hidden Surfaces. Sensors 2018, Vol. 18, Page 2993, 18(9), 2993. https://doi.org/10.3390/S18092993
24. Qayyum, R., Kamal, K., Zafar, T., & Mathavan, S. (2016). Wood defects classification using GLCM based features and PSO trained neural network. 2016 22nd International Conference on Automation and Computing, ICAC 2016: Tackling the New Challenges in Automation and Computing, 273-277. https://doi.org/10.1109/ICONAC.2016.7604931
25. Ren, R., Hung, T., & Tan, K. C. (2018). A Generic Deep-Learning-Based Approach for Automated Surface Inspection. IEEE Transactions on Cybernetics, 48(3), 929-940. https://doi.org/10.1109/TCYB.2017.2668395
26. Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature 1986 323:6088, 323(6088), 533-536. https://doi.org/10.1038/323533a0
27. Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., Berg, A. C., & Fei-Fei, L. (2015). ImageNet Large Scale Visual Recognition Challenge. International Journal of Computer Vision, 115(3), 211-252. https://doi.org/10.1007/S11263-015-0816-Y/FIGURES/16
28. Ruz, G. A., & Estévez, P. A. (2005). Image segmentation using fuzzy min-max neural networks for wood defect detection. Intelligent Production Machines and Systems-First I* PROMS Virtual Conference: Proceedings and CD-ROM Set, 183-189. https://www.researchgate.net/publication/236586295
29. S. Shahnorbanun, S.A.Siti Nurul Huda, A. Haslina, O. Nazlia, & H. Rosilah. (2010). A Computational Biological Network for Wood Defect Classification. Proceedings of the World Congress on Engineering and Computer Science, 559-563. http://iaeng.org/publication/WCECS2010/WCECS2010_pp559-563.pdf
30. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L. C. (2018). MobileNetV2: Inverted Residuals and Linear Bottlenecks. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 4510-4520. https://doi.org/10.1109/CVPR.2018.00474
31. Ser, G., & Bati, C. T. (2019). Derin Sinir Ağları ile En İyi Modelin Belirlenmesi: Mantar Verileri Üzerine Keras Uygulaması. Yüzüncü Yıl Üniversitesi Tarım Bilimleri Dergisi, 29(3), 406-417. https://doi.org/10.29133/YYUTBD.505086
32. Silvén, O., Niskanen, M., & Kauppinen, H. (2003). Wood inspection with non-supervised clustering. Machine Vision and Applications 2003 13:5, 13(5), 275-285. https://doi.org/10.1007/S00138-002-0084-Z
33. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015). Going deeper with convolutions. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 07-12-June-2015, 1-9. https://doi.org/10.1109/CVPR.2015.7298594
34. Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016). Rethinking the Inception Architecture for Computer Vision. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016-December, 2818-2826. https://doi.org/10.1109/CVPR.2016.308
35. Toğaçar, M., Ergen, B., & Özyurt, F. (2020). Evrişimsel Sinir Ağı Modellerinde Özellik Seçim Yöntemlerini Kullanarak Çiçek Görüntülerinin Sınıflandırılması. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 32(1), 47-56. https://doi.org/10.35234/FUMBD.573630
36. Urbonas, A., Raudonis, V., Maskeliunas, R., & Damaševičius, R. (2019). Automated Identification of Wood Veneer Surface Defects Using Faster Region-Based Convolutional Neural Network with Data Augmentation and Transfer Learning. Applied Sciences 2019, Vol. 9, Page 4898, 9(22), 4918. https://doi.org/10.3390/APP9224898
37. Wu, S. Y., Zhang, Z., & Feng, L. (2009). Statistical feature representations for automatic wood defects recognition research and applications. PACIIA 2009 - 2009 2nd Asia-Pacific Conference on Computational Intelligence and Industrial Applications, 1, 19-22. https://doi.org/10.1109/PACIIA.2009.5406462
38. Zhang, X., Zhou, X., Lin, M., & Sun, J. (2018). ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 6848-6856. https://doi.org/10.1109/CVPR.2018.00716