Research Article
BibTex RIS Cite

VGG-16’da Araç Hasar Tespiti

Year 2024, Volume: 5 Issue: 2, 1 - 7
https://doi.org/10.53608/estudambilisim.1421332

Abstract

Günümüzde oluşan trafik kazalarında hızlı bir şekilde hasar tespiti ve buna bağlı olarak hasar kayıtlarının tutulması gerekmektedir. Kazalarda oluşan trafik yoğunluğunu engellemek ve yolu hızlı bir şekilde trafiğe açmak için hasar tespit çalışmalarını hızlandırması önem arz etmektedir. Derin öğrenme teknolojileri hasarın büyüklüğünün hesaplanması, hasar durumunun gösterilmesi ve hasarın maddi boyutu hakkında çıkarımlar yapma konusunda çeşitli avantaj sağlamaktadır. Bu çalışmada sadece sigorta şirketlerinin ya da resmi kurumların sonuçları görmesi için değil, son kullanıcıya da hitap edecek ve oluşan kazaların hasar sınıfını ortaya çıkaracak bir karar destek sistemi amaçlanmıştır. Sunulan yazılım ile sadece kaza süreçlerinde değil, aynı zamanda araç alım-satımı yapılırken hızlı şekilde aracın maddi olarak değerinin belirlenmesinde objektif bir bakış açısı sunmayı amaçlar. Bu çalışmada, CNN alt modeli olan VGG16 tabanlı modelimizi kaggle platformu (5757 adet görüntü) üzerinden elde ettiğimiz veri seti üzerinde eğitim aşamasını geliştirilmiştir. VGG16 ile elde edilen araç nesne tespit oranımız %98, aracın hasarlı olup olmadığının doğruluk oranı %90, hasar oluşan bölgenin tespitini yaptığımız eğitimde elde edilen sonuçlar ise %70 ve son olarak hasarın seviyesini (düşük, orta ve yüksek) belirlediğimiz doğruluk oranı ise %66 olarak elde edilmiştir.

References

  • Çiğdem, A. C. I., & YILMAZ, A. C. (2017). Maddi hasarlı trafik kazaları için sinirsel-bulanık ağ tabanlı bir kusur tespit modeli. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 29(2), 241-250.
  • Zhu, X. J. (2005). Semi-supervised learning literature survey.
  • Nigam, K., McCallum, A. K., Thrun, S., & Mitchell, T. (2000). Text classification from labeled and unlabeled documents using EM. Machine learning, 39, 103-134.
  • Fung, G. P. C., Yu, J. X., Lu, H., & Yu, P. S. (2005). Text classification without negative examples revisit. IEEE transactions on Knowledge and Data Engineering, 18(1), 6-20.
  • Torrey, L., & Shavlik, J. (2010). Transfer learning. In Handbook of research on machine learning applications and trends: algorithms, methods, and techniques (pp. 242-264). IGI global.
  • Weiss, K. R., & Khoshgoftaar, T. M. (2016, November). An investigation of transfer learning and traditional machine learning algorithms. In 2016 IEEE 28th International Conference on Tools with Artificial Intelligence (ICTAI) (pp. 283-290). IEEE.
  • Kyu, P. M., & Woraratpanya, K. (2020, July). Car damage detection and classification. In Proceedings of the 11th international conference on advances in information technology (pp. 1-6).
  • Wang, X., Li, W., & Wu, Z. (2023). Cardd: A new dataset for vision-based car damage detection. IEEE Transactions on Intelligent Transportation Systems.
  • Shirode, A., Rathod, T., Wanjari, P., & Halbe, A. (2022, February). Car damage detection and assessment using CNN. In 2022 IEEE Delhi Section Conference (DELCON) (pp. 1-5). IEEE.
  • Kyu, P. M., & Woraratpanya, K. (2020, July). Car damage detection and classification. In Proceedings of the 11th international conference on advances in information technology (pp. 1-6).
  • Dhieb, N., Ghazzai, H., Besbes, H., & Massoud, Y. (2019, December). A very deep transfer learning model for vehicle damage detection and localization. In 2019 31st international conference on microelectronics (ICM) (pp. 158-161). IEEE.
  • Dwivedi, M., Malik, H. S., Omkar, S. N., Monis, E. B., Khanna, B., Samal, S. R., ... & Rathi, A. (2021). Deep learning-based car damage classification and detection. In Advances in artificial intelligence and data engineering: Select proceedings of AIDE 2019 (pp. 207-221). Springer Singapore.
  • Zhang, Q., Chang, X., & Bian, S. B. (2020). Vehicle-damage-detection segmentation algorithm based on improved mask RCNN. IEEE Access, 8, 6997-7004.
  • Ozdemir, D., & Kunduraci, M. S. (2022). Comparison of deep learning techniques for classification of the insects in order level with mobile software application. IEEE Access, 10, 35675-35684.
  • Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
  • Qassim, H., Verma, A., & Feinzimer, D. (2018, January). Compressed residual-VGG16 CNN model for big data places image recognition. In 2018 IEEE 8th annual computing and communication workshop and conference (CCWC) (pp. 169-175). IEEE.

Car Damage Detection with VGG-16

Year 2024, Volume: 5 Issue: 2, 1 - 7
https://doi.org/10.53608/estudambilisim.1421332

Abstract

In today's traffic accidents, assessing the damage and keeping damage records quickly is necessary. It is crucial to accelerate damage assessment studies to prevent traffic congestion caused by accidents and open the road to traffic quickly. Deep learning technologies provide various advantages in calculating the magnitude of the damage, displaying the damage situation, and making inferences about the material extent of the damage. In this study, a decision support system is aimed not only for insurance companies or official institutions to see the results but also for the end user and to reveal the damage class of the accidents. The software offered aims to provide an objective perspective not only in accident processes but also in quickly determining the financial value of the vehicle when buying and selling the vehicle. In this study, the training phase of our VGG16-based model, a CNN sub-model, was developed on the data set we obtained from the Kaggle platform (5757 images). With VGG16, our vehicle object detection rate is 98%, the accuracy rate of whether the vehicle is damaged is 90%, the results obtained in the training in which we detect the damaged area is 70%, and finally, the accuracy rate in determining the level of damage (low, medium and high) is 66% has been obtained.

References

  • Çiğdem, A. C. I., & YILMAZ, A. C. (2017). Maddi hasarlı trafik kazaları için sinirsel-bulanık ağ tabanlı bir kusur tespit modeli. Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 29(2), 241-250.
  • Zhu, X. J. (2005). Semi-supervised learning literature survey.
  • Nigam, K., McCallum, A. K., Thrun, S., & Mitchell, T. (2000). Text classification from labeled and unlabeled documents using EM. Machine learning, 39, 103-134.
  • Fung, G. P. C., Yu, J. X., Lu, H., & Yu, P. S. (2005). Text classification without negative examples revisit. IEEE transactions on Knowledge and Data Engineering, 18(1), 6-20.
  • Torrey, L., & Shavlik, J. (2010). Transfer learning. In Handbook of research on machine learning applications and trends: algorithms, methods, and techniques (pp. 242-264). IGI global.
  • Weiss, K. R., & Khoshgoftaar, T. M. (2016, November). An investigation of transfer learning and traditional machine learning algorithms. In 2016 IEEE 28th International Conference on Tools with Artificial Intelligence (ICTAI) (pp. 283-290). IEEE.
  • Kyu, P. M., & Woraratpanya, K. (2020, July). Car damage detection and classification. In Proceedings of the 11th international conference on advances in information technology (pp. 1-6).
  • Wang, X., Li, W., & Wu, Z. (2023). Cardd: A new dataset for vision-based car damage detection. IEEE Transactions on Intelligent Transportation Systems.
  • Shirode, A., Rathod, T., Wanjari, P., & Halbe, A. (2022, February). Car damage detection and assessment using CNN. In 2022 IEEE Delhi Section Conference (DELCON) (pp. 1-5). IEEE.
  • Kyu, P. M., & Woraratpanya, K. (2020, July). Car damage detection and classification. In Proceedings of the 11th international conference on advances in information technology (pp. 1-6).
  • Dhieb, N., Ghazzai, H., Besbes, H., & Massoud, Y. (2019, December). A very deep transfer learning model for vehicle damage detection and localization. In 2019 31st international conference on microelectronics (ICM) (pp. 158-161). IEEE.
  • Dwivedi, M., Malik, H. S., Omkar, S. N., Monis, E. B., Khanna, B., Samal, S. R., ... & Rathi, A. (2021). Deep learning-based car damage classification and detection. In Advances in artificial intelligence and data engineering: Select proceedings of AIDE 2019 (pp. 207-221). Springer Singapore.
  • Zhang, Q., Chang, X., & Bian, S. B. (2020). Vehicle-damage-detection segmentation algorithm based on improved mask RCNN. IEEE Access, 8, 6997-7004.
  • Ozdemir, D., & Kunduraci, M. S. (2022). Comparison of deep learning techniques for classification of the insects in order level with mobile software application. IEEE Access, 10, 35675-35684.
  • Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
  • Qassim, H., Verma, A., & Feinzimer, D. (2018, January). Compressed residual-VGG16 CNN model for big data places image recognition. In 2018 IEEE 8th annual computing and communication workshop and conference (CCWC) (pp. 169-175). IEEE.
There are 16 citations in total.

Details

Primary Language Turkish
Subjects Computer Software
Journal Section Research Articles
Authors

Ayberk Gezer 0009-0000-8716-4697

Tolga Yılmaz 0009-0003-4735-1166

Durmuş Özdemir 0000-0002-9543-4076

Early Pub Date July 29, 2024
Publication Date
Submission Date January 17, 2024
Acceptance Date May 9, 2024
Published in Issue Year 2024 Volume: 5 Issue: 2

Cite

IEEE A. Gezer, T. Yılmaz, and D. Özdemir, “VGG-16’da Araç Hasar Tespiti”, Journal of ESTUDAM Information, vol. 5, no. 2, pp. 1–7, 2024, doi: 10.53608/estudambilisim.1421332.

Journal of ESTUDAM Information is indexed by Index Copernicus, Google ScholarASOS Index and ROAD index.