Benchmarking YOLOv8 Against YOLOv5 for Pavement Distress Detection in High-Resolution UAV Imagery: A Photogrammetric Analysis

Document Type : Regular Paper

Authors

1 Civil Engineering Department, Faculty of Engineering, Al-Azhar University, Cairo, Egypt.

2 Civil Engineering Department, Faculty of Engineering, Al-Azhar University, Cairo, Egypt

10.22075/jrce.2025.2415

Abstract

Pavement distress detection is a critical task for ensuring road safety and maintaining transportation infrastructure, particularly in environments with limited resources. This study conducts a comprehensive evaluation of ten YOLO object detection modelscomprising five YOLOv5 and five YOLOv8 variants (n, s, m, l, x) for identifying seven pavement defect classes using 500 high-resolution UAV images. The dataset, manually annotated and split into 70% training and 30% validation sets, was used to train all models under uniform hyperparameter settings. The performance was assessed using standard metrics: mAP@0.5:0.95, mAP@0.5, Recall, Precision and F1-score. Results showed that YOLOv8 consistently outperformed YOLOv5 across all dataset sizes, with YOLOv8l reaching the highest mAP@0.5:0.95 of 0.381, and YOLOv8m providing the best balance between accuracy (0.344), training time (67 minutes), and robustness. Statistical validation using Two-Way ANOVA confirmed significant performance differences (F = 13.81, p = 0.0006, Cohen’s d = 0.73). Further analysis using Repeated Measures ANOVA and Bonferroni-corrected t-tests reinforced YOLOv8l's superiority over other variants. Despite the limited dataset size the findings demonstrate YOLOv8’s effectiveness and reliability in low-resource conditions without the need for image augmentation or preprocessing. The main innovation lies in providing a statistically validated benchmark for UAV-based pavement monitoring under small dataset conditions, highlighting YOLOv8’s superior efficiency and generalization without augmentation or preprocessing. This study provides a reproducible benchmark for real-time UAV-based pavement distress detection, offering insights for deploying lightweight deep learning systems in resource-constrained settings.

Graphical Abstract

Benchmarking YOLOv8 Against YOLOv5 for Pavement Distress Detection in High-Resolution UAV Imagery: A Photogrammetric Analysis

Highlights

  • UAV-based YOLOv8 outperforms YOLOv5 in pavement distress detection accuracy
  • First statistical validation of YOLO models under limited dataset conditions
  • ANOVA tests confirm YOLOv8 robustness without data augmentation or preprocessing
  • UAV + deep learning provides cost-efficient large-scale pavement monitoring

Keywords

Main Subjects

References

[1]      Sekimoto DA and HM and SKG and DT and AM and TK and Y. Deep learning-based road damage detection and classification for multiple countries. Autom Constr 2021;132:103935. https://doi.org/10.1016/j.autcon.2021.103935.
[2]      Zahran Badr, Mohamed El Gendy MER. Using Artificial Neural Networks to Predict the Distresses of Flexible Pavement. J Al-Azhar Univ Eng Sect 2022;17:17–25. https://doi.org/10.21608/auej.2022.216572.
[3]      Marwan Elsayed; H. Mahdi; K. Kandil. Investigating the Factors Affecting Pavement Overlay Service Life. Al-Azhar Univ Civ Eng Res Mag 2019;41:152–64.
[4]      A. A. Rabah ASEHMMR and MEG. A New Revised Version Predictive Model of Elastic Modulus Using Artificial Neural Network. Al-Azhar Univ Civ Eng Res Mag 2021;43:285–96.
[5]      B MK a, B MG, B OR, C BD. Signal processing methodology for detection and localization of damages in columns under the effect of axial load. Measurement 2023;211. https://doi.org/10.1016/j.measurement.2023.112595.
[6]      Fakharian P, Naderpour H. Damage Severity Quantification Using Wavelet Packet Transform and Peak Picking Method. Pract Period Struct Des Constr 2022;27:4021063. https://doi.org/10.1061/(asce)sc.1943-5576.0000639.
[7]      A BM, B MK, B OR, B MG, Gholamreza Ghodrati Amiri a. Efficient modal flexibility signal processing methodology for innovative identification of debonding zones in concrete-filled steel tubes. Results Eng 2025;26. https://doi.org/10.1016/j.rineng.2025.104683.
[8]      Khanahmadi M. A cutting-edge framework for damage-sensitive feature extraction leveraging modal dynamic flexibility in signal processing-driven structural health monitoring. Thin-Walled Struct 2025;216. https://doi.org/10.1016/j.tws.2025.113617.
[9]      Naderpour H, Fakharian P. A synthesis of peak picking method and wavelet packet transform for structural modal identification. KSCE J Civ Eng 2016;20:2859–67. https://doi.org/10.1007/s12205-016-0523-4.
[10]     Zhu J. Pavement distress detection using convolutional neural networks with images captured via UAV. Autom Constr 2022. https://doi.org/10.1016/j.autcon.2021.103991.
[11]     Samadzadegan F, Dadrass Javan F, Ashtari Mahini F, Gholamshahi M, Nex F. Automatic road pavement distress recognition using deep learning networks from unmanned aerial imagery. Drones 2024;8:244. https://doi.org/10.3390/drones8060244.
[12]     Zhu J et al. Pavement distress detection using convolutional neural networks with images captured via UAV. Autom Constr 2022;133:103991. https://doi.org/10.1016/j.autcon.2021.103991.
[13]     Yan K, Zhang Z. Automated asphalt highway pavement crack detection based on deformable single shot multi-box detector under a complex environment. IEEE Access 2021;9:150925–150938. https://doi.org/10.1109/ACCESS.2021.3125703.
[14]     Song L, Wang X. Faster region convolutional neural network for automated pavement distress detection. Road Mater Pavement Des 2021;22:23–41. https://doi.org/10.1080/14680629.2019.1614969.
[15]     Silva LA, Sanchez San Blas H, Peral García D, Sales Mendes A, Villarubia González G. An architectural multi-agent system for a pavement monitoring system with pothole recognition in UAV images. Sensors 2020;20:6205. https://doi.org/10.3390/s20216205.
[16]     Shaghouri AA, Alkhatib R, Berjaoui S. Real-time pothole detection using deep learning. ArXiv 2021. https://doi.org/10.48550/arXiv.2107.06356.
[17]     Shu Z, Yan Z, Xu X. Pavement crack detection method of street view images based on deep learning. J Phys Conf Ser 2021;1952:022043. https://doi.org/10.1088/1742-6596/1952/2/022043.
[18]     Park S-S, Tran V-T, Lee D-E. Application of various yolo models for computer vision-based real-time pothole detection. Appl Sci 2021;11:11229. https://doi.org/10.3390/app112311229.
[19]     Hu GX, Hu BL, Yang Z, Huang Y, Li P. Pavement crack detection method based on deep learning models. Wirel Commun Mob Comput 2021;2021:5573590. https://doi.org/10.1155/2021/5573590.
[20]     Xu X et al. Crack detection and comparison study based on Faster R-CNN and Mask R-CNN. Sensors 2022;22. https://doi.org/10.3390/s22031215.
[21]     Guo L, Li R, Jiang B. A road surface damage detection method using YOLOv4 with PID optimizer. Int J Innov Comput Inf Control 2021;17:1763–1774. https://doi.org/10.24507/ijicic.17.05.1763.
[22]     El-Sayed MFA and MAH and ES and AAE-S and RFA-K and AM and E. EGY_PDD: a comprehensive multi-sensor benchmark dataset for accurate pavement distress detection and classification. Multimed Tools Appl 2025. https://doi.org/10.1007/s11042-025-20700-w.
[23]     Wang LF and ST and MKA b. MA and MISI and X. Impact of Image Preprocessing and Crack Type Distribution on YOLOv8-Based Road Crack Detection. Sensors 2025;25:2180. https://doi.org/10.3390/s25072180.
[24]     Zhang XC and CW and CL and XZ and YZ and TL and J. Autonomous Crack Detection for Mountainous Roads Using UAV Inspection System. Sensors 2024;24:4751. https://doi.org/10.3390/s24094751.
[25]     Guo JZ and HX and PL and KZ and WH and R. Pavement Crack Detection Method via Deep Learning and a Binocular-Vision-Based Unmanned Aerial Vehicle. Appl Sci 2024;14:1778. https://doi.org/10.3390/app14051778.
[26]     Lin YTJ and HTY and YRZ and KQW and RYL and CY. RDD-YOLOv5: Road Defect Detection Algorithm with Self-Attention Based on Unmanned Aerial Vehicle Inspection. Sensors 2023;23:8241. https://doi.org/10.3390/s23198241.
[27]     Wu XX and HH and YD and YZ and S. GC-YOLOv5s: A Lightweight Detector for UAV Road Crack Detection. Appl Sci 2023;13:11030. https://doi.org/10.3390/app131911030.
[28]     Li XW and HG and ZJ and Z. BL-YOLOv8: An Improved Road Defect Detection Model Based on YOLOv8. Sensors 2023;23:8361. https://doi.org/10.3390/s23208361.
[29]     Ultralytics. , YOLOv8. Available:Https://GithubCom/Ultralytics/Ultralytics 2025.
[30]     Huang GW and YC and PA and HH and JH and T. UAV-YOLOv8: A Small-Object-Detection Model Based on Improved YOLOv8 for UAV Aerial Photography Scenarios. Sensors 2023;23:7190. https://doi.org/10.3390/s23167190.
[31]     Wang XC and CL and LC and XZ and YZ and C. A Pavement Crack Detection and Evaluation Framework for a UAV Inspection System Based on Deep Learning. Appl Sci 2024;14:1157. https://doi.org/10.3390/app14031157.
[32]     Nouri Y, Shahabian F, Shariatmadar H, Entezami A. Structural Damage Detection in the Wooden Bridge Using the Fourier Decomposition, Time Series Modeling and Machine Learning Methods. J Soft Comput Civ Eng 2024;8:83–101. https://doi.org/10.22115/scce.2023.401971.1669.
[33]     Soleimani Nezhad, S., Khademian, F., Naderpour, H., Kalantari, S. M., & Fakharian P. Signal Processing-Based Damage Detection of Steel Braced Frame Subjected to Consequent Excitations. Innov Infrastruct Solut 2024;9. https://doi.org/10.1007/s41062-024-01762-5.
[34]     A HRA, B NM, Mahmoud Bayat c. A novel damage identification method based on short time Fourier transform and a new efficient. Structures 2021;33:3605–14. https://doi.org/10.1016/j.istruc.2021.06.081.
[35]     Agüera-Vega F. C-RF. M-CP. Accuracy of digital surface models and orthophotos derived from unmanned aerial vehicle photogrammetry. J Surv Eng 2017;143:04016025. https://doi.org/10.1061/(ASCE)SU.1943-5428.0000206.
[36]     Pádua L. MP. HJ. AT. PE. MR. SJJ. UAV-Based Forest Fire Detection and Monitoring Using Multi-Temporal and Multi-Sensor Data. Remote Sens 2022;14:983. https://doi.org/10.3390/rs14040983.
[37]     Redmon J. You Only Look Once. Proc. IEEE Conf. Comput. Vis. Pattern Recognit., 2016.
[38]     Astrup SP and R. Automatic detection of snow breakage at single tree level using YOLOv5 applied to UAV imagery. Int J Appl Earth Obs Geoinf 2022;112:102946. https://doi.org/10.1016/j.jag.2022.102946.
[39]     Lee JD and YL and VCS. Concrete crack detection with handwriting script interferences using faster region-based convolutional neural network. Comput Aided Civ Infrastruct Eng 2020;35:373–388. https://doi.org/10.1111/mice.12497.
[40]     Wilson AC. RR. SM. SN. RB. The marginal value of adaptive gradient methods in machine learning. Adv Neural Inf Process Syst 2017;30:4148–58. https://doi.org/10.48550/arXiv.1705.08292.
[41]     E. Fawaz. Statistics lectures : ANOVA‑test 2022.

Files

History

  • Receive Date: 08 August 2025
  • Revise Date: 14 September 2025
  • Accept Date: 21 November 2025

Share

How to cite

Statistics