Optimizing YOLO Models for Enhanced Road Damage Detection: A Performance Comparison of YOLOv5 and YOLOv8
DOI:
https://doi.org/10.21831/elinvo.v10i2.88919Keywords:
YOLOv5, YOLOv8, Deep learning, GRDDC 2020 Dataset, Object detectionAbstract
Accurate road damage detection is vital for ensuring road safety and infrastructure maintenance. This study evaluates and compares the performance of four YOLO models—YOLOv5-S, YOLOv5-M, YOLOv8-S, and YOLOv8-M—for detecting road damage types such as Alligator Cracks, Longitudinal Cracks, Transverse Cracks, Potholes, and Lateral Cracks. The models were trained on a combined dataset from GRDDC 2020 and the Ministry of Public Works and Housing (PUPR) Republic of Indonesia, addressing challenges like class imbalance and diverse road conditions. Results show that YOLOv8-M achieved the highest mAP@0.5 (0.412), excelling in precision and recall for prominent damage types, making it the most reliable for high-accuracy applications. YOLOv5-M balanced precision and recall, while YOLOv5-S prioritized recall, making it suitable for detecting widespread damage. However, all models struggled with less prominent types, such as Lateral Cracks, due to class imbalance. Misclassifications were common, with the "Background" class absorbing predictions from other categories. This study highlights the strengths and limitations of each model, offering insights into improving road damage detection through better feature extraction, expanded datasets, and optimized architectures. These findings provide a foundation for deploying automated deep learning-based road damage detection systems to enhance infrastructure management.
References
[1] B. P. S.-S. Indonesia, “Panjang Jalan Menurut Provinsi dan Tingkat Kewenangan Pemerintahan (km),” in Last modified August 13. [Online]. Available: https://www.bps.go.id/en/statistics-table/3/U0VOeFZEZFNiVnByUkdGMlNrOTFVVGRHY1ZkVGR6MDkjMw==/panjang-jalan-menurut-provinsi-dan-tingkat-kewenangan-pemerintahan–km—2022.html?year=2023.
[2] P. J. Gertler, M. Gonzalez-Navarro, T. Gračner, and A. D. Rothenberg, “Road maintenance and local economic development: Evidence from Indonesia’s highways,” J Urban Econ, vol. 143, p. 103687, Sep. 2024, doi: 10.1016/j.jue.2024.103687.
[3] B. P. S.-S. Indonesia, “Traffic Accident, Killed Person, Seriously Injured, Slight Injured and Expected of Material Losses Value.” [Online]. Available: https://www.bps.go.id/en/statistics-table/2/NTEzIzI%3D/traffic-accident–killed-person–seriously-injured–slight-injured-and-expected-of-material-losses-value.html.
[4] D. Willar, A. A. D. P. Dewi, and F. P. Makalew, “Reviewing Quality Control Management of Road Construction Projects,” in Proceedings of the 5th International Conference on Sustainable Civil Engineering Structures and Construction Materials (SCESCM 2020, vol. 215, S. Belayutham, C. K. I. C. Ibrahim, A. Alisibramulisi, H. Mansor, and M. Billah, Eds., Singapore: Springer, 2022, pp. 1261–1271. doi: 10.1007/978-981-16-7924-7_82.
[5] Y. Sari and M. H. Yudhistira, “Bad light, bad road, or bad luck? The associations of road lighting and road surface quality on road crash severities in Indonesia,” Case Stud Transp Policy, vol. 9, no. 3, pp. 1407–1417, Sep. 2021, doi: 10.1016/j.cstp.2021.07.014.
[6] P. Jiang, D. Ergu, F. Liu, Y. Cai, and B. Ma, “A Review of Yolo Algorithm Developments,” Procedia Comput Sci, vol. 199, pp. 1066–1073, 2022, doi: 10.1016/j.procs.2022.01.135.
[7] D. Arya et al., “Deep learning-based road damage detection and classification for multiple countries,” Autom Constr, vol. 132, p. 103935, Dec. 2021, doi: 10.1016/j.autcon.2021.103935.
[8] Norsuzila Ya’acob, Mohamad Danial Ikmal Zuraimi, Amirul Asraf Abdul Rahman, Azita Laily Yusof, and Darmawaty Mohd Ali, “Real-Time Pavement Crack Detection Based on Artificial Intelligence,” Journal of Advanced Research in Applied Sciences and Engineering Technology, vol. 38, no. 2, pp. 71–82, Jan. 2024, doi: 10.37934/araset.38.2.7182.
[9] Z. Diao, X. Huang, H. Liu, and Z. Liu, “LE‐YOLOv5: A Lightweight and Efficient Road Damage Detection Algorithm Based on Improved YOLOv5,” International Journal of Intelligent Systems, vol. 2023, no. 1, Jan. 2023, doi: 10.1155/2023/8879622.
[10] M. Khan, M. A. Raza, G. Abbas, S. Othmen, A. Yousef, and T. A. Jumani, “Pothole detection for autonomous vehicles using deep learning: a robust and efficient solution,” Front Built Environ, vol. 9, p. 1323792, Jan. 2024, doi: 10.3389/fbuil.2023.1323792.
[11] N. Chitraningrum et al., “Comparison Study of Corn Leaf Disease Detection based on Deep Learning YOLO-v5 and YOLO-v8,” Journal of Engineering and Technological Sciences, vol. 56, no. 1, pp. 61–70, Feb. 2024, doi: 10.5614/j.eng.technol.sci.2024.56.1.5.
[12] D. Arya et al., “Global Road Damage Detection: State-of-the-art Solutions,” in 2020 IEEE International Conference on Big Data (Big Data), IEEE, Dec. 2020, pp. 5533–5539. doi: 10.1109/BigData50022.2020.9377790.
[13] G. Jocher, “Ultralytics/YOLOv5: V6.0—YOLOv5n ‘Nano’ Models, Roboflow Integration, TensorFlow Export, OpenCV DNN Support,” Zenodo.
[14] A. Rácz, D. Bajusz, and K. Héberger, “Effect of Dataset Size and Train/Test Split Ratios in QSAR/QSPR Multiclass Classification,” Molecules, vol. 26, no. 4, p. 1111, Feb. 2021, doi: 10.3390/molecules26041111.
[15] J. Miao and W. Zhu, “Precision–recall curve (PRC) classification trees,” Evol Intell, vol. 15, no. 3, pp. 1545–1569, Sep. 2022, doi: 10.1007/s12065-021-00565-2.
[16] A. A. Sami, S. Sakib, K. Deb, and I. H. Sarker, “Improved YOLOv5-Based Real-Time Road Pavement Damage Detection in Road Infrastructure Management,” Algorithms, vol. 16, no. 9, p. 452, Sep. 2023, doi: 10.3390/a16090452.
[17] Q.-H. Phan, V.-T. Nguyen, C.-H. Lien, T.-P. Duong, M. T.-K. Hou, and N.-B. Le, “Classification of Tomato Fruit Using Yolov5 and Convolutional Neural Network Models,” Plants, vol. 12, no. 4, p. 790, Feb. 2023, doi: 10.3390/plants12040790.
[18] J. Torres, “YOLOv8 Architecture.” [Online]. Available: https://yolov8.org/yolov8-architecture/.
[19] K. Indra Sari, B. Sugiarto Waloejo, and I. Widyawati Agustin, “Comparative Study: Level of Service in Indonesia and India,” International Journal of Science and Research (IJSR), vol. 11, no. 7, pp. 1333–1337, Jul. 2022, doi: 10.21275/SR22718055914.
[20] A. Kanoungo, U. Sharma, A. Goyal, S. Kanoungo, and S. Singh, “Assessment of Causes of Pothole Development on Chandigarh Roads,” Journal of The Institution of Engineers (India): Series A, vol. 102, no. 2, pp. 411–419, Jun. 2021, doi: 10.1007/s40030-021-00520-5.
[21] D. Arya, H. Maeda, S. K. Ghosh, D. Toshniwal, and Y. Sekimoto, “RDD2020: An annotated image dataset for automatic road damage detection using deep learning,” Data Brief, vol. 36, p. 107133, Jun. 2021, doi: 10.1016/j.dib.2021.107133.
[22] L. P. Ingrassia, P. Spinelli, G. Paoloni, and F. Canestrari, “Top-down cracking in Italian motorway pavements: A case study,” Case Studies in Construction Materials, vol. 13, p. e00442, Dec. 2020, doi: 10.1016/j.cscm.2020.e00442.
[23] Y. Wang, A. W. Z. Chew, and L. Zhang, “Building damage detection from satellite images after natural disasters on extremely imbalanced datasets,” Autom Constr, vol. 140, p. 104328, Aug. 2022, doi: 10.1016/j.autcon.2022.104328.
[24] M. R. Rezaei-Dastjerdehei, A. Mijani, and E. Fatemizadeh, “Addressing Imbalance in Multi-Label Classification Using Weighted Cross Entropy Loss Function,” in 2020 27th National and 5th International Iranian Conference on Biomedical Engineering (ICBME), IEEE, Nov. 2020, pp. 333–338. doi: 10.1109/ICBME51989.2020.9319440.
[25] T. Saito and M. Rehmsmeier, “The Precision-Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets,” PLoS One, vol. 10, no. 3, p. e0118432, Mar. 2015, doi: 10.1371/journal.pone.0118432.
[26] J. Davis and M. Goadrich, “The relationship between Precision-Recall and ROC curves,” in Proceedings of the 23rd international conference on Machine learning - ICML ’06, New York, New York, USA: ACM Press, 2006, pp. 233–240. doi: 10.1145/1143844.1143874.
[27] P. Branco, L. Torgo, and R. Ribeiro, “A Survey of Predictive Modelling under Imbalanced Distributions,” May 2015, [Online]. Available: http://arxiv.org/abs/1505.01658
[28] V. Mandal, A. R. Mussah, and Y. Adu-Gyamfi, “Deep Learning Frameworks for Pavement Distress Classification: A Comparative Analysis,” in 2020 IEEE International Conference on Big Data (Big Data), IEEE, Dec. 2020, pp. 5577–5583. doi: 10.1109/BigData50022.2020.9378047.
[29] “YOLOv5 Architecture.” [Online]. Available:https://github.com/ultralytics/yolov5/issues/280
[30] R. Padilla, W. L. Passos, T. L. B. Dias, S. L. Netto, and E. A. B. da Silva, “A comparative analysis of object detection metrics with a companion open-source toolkit,” Electronics, vol. 10, no. 3, Art. no. 279, 2021, doi: 10.3390/electronics10030279.
Downloads
Published
How to Cite
Issue
Section
Citation Check
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The article published in ELINVO became ELINVO's right in publication.
This work by ELINVO is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
