YOLO: A Competitive Analysis of Modern Object Detection Algorithms for Road Defects Detection Using Drone Images
Main Article Content
Abstract
Efficient identification of road defects is a critical concern for road safety and infrastructure upkeep. This research employs drone-captured imagery and advanced object detection algorithms to expedite defect recognition, with a specific focus on determining the optimal algorithm for prompt and precise detection. The importance of timely road defect detection, crucial for mitigating potential hazards, remains central. A comprehensive comparative analysis of contemporary object detection algorithms, encompassing YOLOv5s, YOLOv5m, YOLOv5l, YOLOv5x, and YOLOv7. The results of this study highlight YOLOv7 as the most efficient, with a notable mAP of 68.3%, closely followed by YOLOv5l (66.8%), YOLOv5m (66.3%), YOLOv5x (66%), and YOLOv5s (63%). The integration of drone-derived imagery, capturing distinct gradients, significantly enhances defect detection accuracy. Beyond road safety, this study offers valuable insights to computer vision and machine learning practitioners. By bridging technological innovation with practical implementation, it holds potential to advance road safety and transportation infrastructure quality and the use of revolutionary drone technology.
Received 04/05/2023,
Revised 10/10/2023y,
Accepted 10/10/2023,
Published Online First 20/11/2023
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite
References
Seo J, Duque L, Wacker J. Drone-enabled bridge inspection methodology and application. Autom Constr. 2018; 94: 112-126. https://doi.org/10.1016/j.autcon.2018.06.006
Ha J, Kim DS, Kim M. Assessing severity of road cracks using deep learning-based segmentation and detection. J Supercomput. 2022 May 22; 78(16): 17721–35. https://doi.org/10.1007/s11227-022-04560-x
Jang J, Yang Y, Smyth AW, Cavalcanti D, Kumar R. Framework of data acquisition and Integration for the detection of pavement distress via multiple vehicles. J Comput Civil Eng. 2017 Mar 1; 31(2): 04016061. https://doi.org/10.1061/(asce)cp.1943-5487.0000618
Syifa M, Park S, Lee CW. Detection of the Pine wilt disease tree candidates for drone remote sensing using artificial intelligence techniques. Eng. 2020 Aug 1; 6(8): 919–26. https://doi.org/10.1016/j.eng.2020.07.001
Mandirola M, Casarotti C, Peloso S, Lanese I, Brunesi E, Senaldi I. Use of UAS for damage inspection and assessment of bridge infrastructures. Int J Disaster Risk Reduct. 2022 Apr 1; 72: 102824. https://doi.org/10.1016/j.ijdrr.2022.102824
Awad JH, Majeed BD. Moving objects detection based on frequency domain. Baghdad Sci J. 2020 May 11; 17(2): 0556. https://doi.org/10.21123/bsj.2020.17.2.0556
Balakrishnan B, Chelliah R, Venkatesan M, Sah C. Comparative Study on Various Architectures of Yolo Models Used In Object Recognition. 2022 International Conference on Computing, Communication, and Intelligent Systems (ICCCIS). 2022 Nov 4; https://doi.org/10.1109/icccis56430.2022.10037635
Yusro MM, Ali R, Hitam MS. Comparison of faster R-CNN and YOLOV5 for overlapping objects recognition. Baghdad Sci J. 2023; 20(3): 893-903. https://doi.org/10.21123/bsj.2022.7243
Wang Z, Zhu H, Jia X, Bao Y, Wang C. Surface Defect Detection with Modified Real-Time Detector YOLOv3. J Sens. 2022; 2022. https://doi.org/10.1155/2022/8668149
Redmon J, Santosh DHH, Ross G, Farhadi A. You Only Look Once: Unified, Real-Time Object Detection. arXiv (Cornell University). 2015 Jun 8. http://arxiv.org/abs/1506.02640
Norkobil Saydirasulovich S, Abdusalomov A, Jamil MK, Nasimov R, Kozhamzharova D, Cho YI. A YOLOv6-Based Improved Fire Detection Approach for Smart City Environments. Sensors. 2023; 23(6): 3161. https://doi.org/10.3390/s23063161
Azurmendi I, Zulueta E, López-Guede JM, Azkarate J, González M. Cooktop sensing based on a YOLO object detection algorithm. Sensors. 2023 Mar 3; 23(5): 2780: https://doi.org/10.3390/s23052780
Dluznevskij D, Stefanovič P, Ramanauskaitė S. Investigation of YOLOV5 efficiency in iPhone supported systems. Balt J Mod Comput. 2021 Jan 1;9(3):07. https://doi.org/10.22364/bjmc.2021.9.3.07
Wang CY, Bochkovskiy A, Liao HYM. YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. arXiv (Cornell University). 2022 Jul 6; 9(3). http://arxiv.org/abs/2207.02696
Pham V, Pham C, Dang T. Road Damage Detection and Classification with Detectron2 and Faster R-CNN. Proceedings - 2020 IEEE Int Conf Big Data. Published online October 28, 2020: 5592-5601. https://doi.org/10.1109/BigData50022.2020.9378027
Tan M, Pang R, Le Q V. EfficientDet: Scalable and Efficient Object Detection. 2020 IEEE/CVF Conf Comp Vis Pattern Recognition (CVPR). Published online 2019: 10778-10787. https://doi.org/10.1109/CVPR42600.2020.01079
Vishwakarma R, Vennelakanti R. CNN Model & Tuning for Global Road Damage Detection. Proceedings - 2020 IEEE Int Conf Big Data. 2020. Published online March 17, 2021: 5609-5615. https://doi.org/10.1109/BigData50022.2020.9377902
Arya D, Maeda H, Ghosh SK, Toshniwal D, Omata H, Kashiyama T, et al. Global Road Damage Detection: State-of-the-art Solutions. Proceedings - 2020 IEEE Int Conf Big Data. Published online November 17, 2020: 5533-5539. https://doi.org/10.1109/BigData50022.2020.9377790