DERİN ÖĞRENME İLE ASFALT ÇATLAKLARININ TESPİTİNDE VERİ ARTIRIMI VE EVRİŞİMSEL BLOK SEÇİMİNİN ETKİSİ

Year 2024, Volume: 11 Issue: 23, 172 - 189, 31.08.2024

Zahide Topbaş , Özlem Erdaş Çiçek , Şaban Gülcü

https://doi.org/10.54365/adyumbd.1453460

Abstract

Petrolün rafine işlemi sonucu yan bir ürün olarak oluşan asfalt, yol çalışmalarında kullanılan en önemli malzemelerdendir. Asfalt yollarda trafik, iklim ve çevre, yapım ve tasarım hataları, malzeme hataları gibi etkenlerden dolayı deformasyonlar, ayrışmalar ve çatlaklar meydana gelmektedir. Bu bozulmalar asfalt yolun kalitesini düşürmekte ve kazalara sebep olmaktadır. Bu çalışma, onarım ve altyapı iyileştirmelerinin sağlanması için çatlaklar gibi asfalt hasarlarının tespitini iyileştirmeyi amaçlamaktadır. Çalışma kapsamında, Mendeley Data'nın "Asfalt Çatlak Veri Seti" üzerinde eğitilmiş bir derin öğrenme modelini kullanarak, çatlak görüntüleri %96'yı aşan bir doğruluk oranıyla sınıflandırılmıştır. Bu araştırma, derin sinir ağlarının endüstriyel uygulamalarda kullanımının ürün kalitesini iyileştirme potansiyelini ortaya koymaktadır.

Keywords

Derin öğrenme, görüntü sınıflandırma, evrişimli sinir ağları, asfalt çatlağı tespiti, veri arttırım

THE EFFECT OF DATA AUGMENTATION AND CONVOLUTIONAL BLOCK SELECTION ON THE DETECTION OF ASPHALT CRACKS WITH DEEP LEARNING

Abstract

Asphalt, formed as a byproduct through the refining process of oil, is one of the most crucial materials used in road construction. Deformations, separations, and cracks occur in asphalt roads due to factors such as traffic, climate and environment, construction and design errors, and material defects. These deteriorations reduce the quality of asphalt roads and lead to accidents. This study aims to improve the detection of asphalt damages such as cracks to enable repairs and infrastructure improvements. Within the scope of the study, crack images were classified with an accuracy rate exceeding 96% using a deep learning model trained on Mendeley Data's "Asphalt Crack Data Set". This research demonstrates the potential of using deep neural networks in industrial applications to improve product quality.

Keywords

Deep learning, image classification, convolutional neural networks, asphalt crack detection, data augmentation

References

• B. Kesayak, “Endüstri Tarihine Kısa Bir Yolculuk,” https://www.endustri40.com/endustri-tarihine-kisa-bir-yolculuk/.
• “Yol Ağı Bilgileri,” https://www.kgm.gov.tr/Sayfalar/KGM/SiteTr/Kurumsal/YolAgi.
• “Asfalt Yollarda Meydana Gelen Kusur ve Hasar Çeşitleri,” https://insapedia.com/asfalt-yollarda-meydana-gelen-kusur-ve-hasar-cesitleri.
• A. Erbaş and D. İlek, “Son dakika haberi... Karadeniz Sahil Yolu’nda ‘çatlak’ tehlikesi,” https://www.hurriyet.com.tr/gundem/yolda-catlak-tehlikesi-ekipler-hemen-harekete-gecti-41715204, Rize, 2021.
• A. Şeker, B. Diri, and H. H. Balık, “Derin Öğrenme Yöntemleri ve Uygulamaları Hakkında Bir İnceleme,” Gazi Mühendislik Bilimleri Dergisi , vol. 3, no. 3, pp. 47–64, 2017.
• J. Chadwick, “Autonomous pothole-repairing robots will hit Britain’s streets by 2021,” https://www.dailymail.co.uk/sciencetech/article-8883627/Autonomous-pothole-repairing-robots-hit-Britains-streets-2021.html, 2020.
• M. Bayğın, O. Yaman, and T. Tuncer, “Akıllı Şehirler için Özellik Çıkarımı ve Makine Öğrenmesi Tabanlı Asfalt Durum İzleme Yaklaşımı,” European Journal of Science and Technology, Jan. 2021, doi: 10.31590/ejosat.844592.
• L. de A. Schmidt, M. Pasin, and S. M. Peres, “Feasibility Study for an Automatic Architecture for Pothole Detection in Asphalt Images: a Trade-off between Performance and Quality,” in 2020 IEEE 16th International Conference on Intelligent Computer Communication and Processing (ICCP), IEEE, Sep. 2020, pp. 307–314. doi: 10.1109/ICCP51029.2020.9266229.
• N. Shatnawi, “Automatic Pavement Cracks Detection using Image Processing Techniques and Neural Network,” International Journal of Advanced Computer Science and Applications, vol. 9, no. 9, 2018, doi: 10.14569/IJACSA.2018.090950.
• V. Mandal, L. Uong, and Y. Adu-Gyamfi, “Automated Road Crack Detection Using Deep Convolutional Neural Networks,” in 2018 IEEE International Conference on Big Data (Big Data), IEEE, Dec. 2018, pp. 5212–5215. doi: 10.1109/BigData.2018.8622327.
• J. Cheng, W. Xiong, W. Chen, Y. Gu, and Y. Li, “Pixel-level Crack Detection using U-Net,” in TENCON 2018 - 2018 IEEE Region 10 Conference, IEEE, Oct. 2018, pp. 0462–0466. doi: 10.1109/TENCON.2018.8650059.
• E. Deveci and B. Ergen, “Yol Yüzey Anormalisinin Tespiti ve Analizi,” Bitlis Eren Üniversitesi Fen Bilimleri Dergisi, vol. 10, no. 3, pp. 1187–1194, Sep. 2021, doi: 10.17798/bitlisfen.942386.
• F. Liu, J. Liu, and L. Wang, “Asphalt Pavement Crack Detection Based on Convolutional Neural Network and Infrared Thermography,” IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 11, pp. 22145–22155, Nov. 2022, doi: 10.1109/TITS.2022.3142393.
• Y. Que et al., “Automatic classification of asphalt pavement cracks using a novel integrated generative adversarial networks and improved VGG model,” Eng Struct, vol. 277, p. 115406, Feb. 2023, doi: 10.1016/j.engstruct.2022.115406.
• S. Matarneh, F. Elghaish, F. Pour Rahimian, E. Abdellatef, and S. Abrishami, “Evaluation and optimisation of pre-trained CNN models for asphalt pavement crack detection and classification,” Autom Constr, vol. 160, p. 105297, Apr. 2024, doi: 10.1016/j.autcon.2024.105297.
• M. Yılmaz et al., “Improving the Classification Performance of Asphalt Cracks After Earthquake With a New Feature Selection Algorithm,” IEEE Access, vol. 12, pp. 6604–6614, 2024, doi: 10.1109/ACCESS.2023.3343619.
• A. Jayanth Balaji, G. Thiru Balaji, M. S. Dinesh, N. Binoy, and D. S. Harish Ram, “Asphalt Crack Dataset,” Available: https://data.mendeley.com/datasets/xnzhj3x8v4/2, vol. 2. 2019.
• A. J. Balaji, G. Thiru Balaji, M. S. Dinesh, B. B. Nair, and D. S. Harish Ram, “A Machine Learning Based Approach to Crack Detection in Asphalt Pavements,” in 2018 15th IEEE India Council International Conference (INDICON), IEEE, Dec. 2018, pp. 1–4. doi: 10.1109/INDICON45594.2018.8987039.
• B. Ji, “A Study on Crack Detection in Asphalt Road Pavement Using Small Deep Learning,” Journal of the Korean GEO-environmental Society, vol. 22, no. 10, pp. 13–19, 2021.
• A. Kumar, A. Kumar, A. K. Jha, and A. Trivedi, “Crack Detection of Structures using Deep Learning Framework,” in 2020 3rd International Conference on Intelligent Sustainable Systems (ICISS), IEEE, Dec. 2020, pp. 526–533. doi: 10.1109/ICISS49785.2020.9315949.
• “ImageNet,” https://www.image-net.org/about.php.
• “VGG16 and VGG19,” https://keras.io/api/applications/vgg/.
• T. Fushiki, “Estimation of prediction error by using K-fold cross-validation,” Stat Comput, vol. 21, no. 2, pp. 137–146, Apr. 2011, doi: 10.1007/s11222-009-9153-8.
• C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J Big Data, vol. 6, no. 1, p. 60, Dec. 2019, doi: 10.1186/s40537-019-0197-0.
• M. Toğaçar, K. A. Eşidir, and B. Ergen, “Yapay Zekâ Tabanlı Doğal Dil İşleme Yaklaşımını Kullanarak İnternet Ortamında Yayınlanmış Sahte Haberlerin Tespiti,” Journal of Intelligent Systems: Theory and Applications, vol. 5, no. 1, pp. 1–8, Oct. 2021, doi: 10.38016/jista.950713.
• Ö. İnik and E. Ülker, “Derin Öğrenme ve Görüntü Analizinde Kullanılan Derin Öğrenme Modelleri ,” GAZİOSMANPAŞA BİLİMSEL ARAŞTIRMA DERGİSİ (GBAD), vol. 6, no. 3, pp. 85–104, 2017.
• A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Commun ACM, vol. 60, no. 6, pp. 84–90, May 2017, doi: 10.1145/3065386.
• K. Fırıldak and M. F. Talu, “Evrişimsel Sinir Ağlarında Kullanılan Transfer Öğrenme Yaklaşımlarının İncelenmesi,” Computer Science, vol. 4, no. 2, pp. 88–95, 2019.
• J. Xu, Z. Li, B. Du, M. Zhang, and J. Liu, “Reluplex made more practical: Leaky ReLU,” in 2020 IEEE Symposium on Computers and Communications (ISCC), IEEE, Jul. 2020, pp. 1–7. doi: 10.1109/ISCC50000.2020.9219587.
• U. Ruby, P. Theerthagiri, I. J. Jacob, and Y. Vamsidhar, “Binary cross entropy with deep learning technique for Image classification,” International Journal of Advanced Trends in Computer Science and Engineering, vol. 9, no. 4, pp. 5393–5397, Aug. 2020, doi: 10.30534/ijatcse/2020/175942020.
• M. A. Koşan, A. Coşkun, and H. Karacan, “Yapay Zekâ Yöntemlerinde Entropi,” Journal of Information Systems and Management Research, vol. 1, no. 1, pp. 15–22, 2019.
• G. Ser and C. T. Bati, “Derin Sinir Ağları ile En İyi Modelin Belirlenmesi: Mantar Verileri Üzerine Keras Uygulaması,” Yüzüncü Yıl Üniversitesi Tarım Bilimleri Dergisi, vol. 29, no. 3, pp. 406–417, Sep. 2019, doi: 10.29133/yyutbd.505086.
• R. V. Kumar Reddy, B. Srinivasa Rao, and K. P. Raju, “Handwritten Hindi Digits Recognition Using Convolutional Neural Network with RMSprop Optimization,” in 2018 Second International Conference on Intelligent Computing and Control Systems (ICICCS), IEEE, Jun. 2018, pp. 45–51. doi: 10.1109/ICCONS.2018.8662969.
• H. M and S. M.N, “A Review on Evaluation Metrics for Data Classification Evaluations,” International Journal of Data Mining & Knowledge Management Process, vol. 5, no. 2, pp. 01–11, Mar. 2015, doi: 10.5121/ijdkp.2015.5201.
• E. ERGÜN and K. KILIÇ, “Derin Öğrenme ile Artırılmış Görüntü Seti üzerinden Cilt Kanseri Tespiti,” Black Sea Journal of Engineering and Science, vol. 4, no. 4, pp. 192–200, Oct. 2021, doi: 10.34248/bsengineering.938520.