AN EFFECTIVE APPROACH FOR IMPROVING ROAD DAMAGE DETECTION PERFORMANCE
Abstract
In recent years, convolutional neural network-based object detection models have garnered significant attention and demonstrated impressive performance. Nevertheless, the accuracy of detecting potholes has not met expectations due to the scarcity of adequate datasets for training and evaluation purposes. This study aims to address the challenges associated with pothole detection by introducing a novel dataset named PH. The PH dataset comprises 2,612 real-world pothole images with bounding box annotations, depicting diverse driving scenes in various weather conditions. The PH dataset is employed to train three state-of-the-art object detection models, including SSD512, RetinaNet, and YOLOv7. Experimental findings
indicate that the models achieve pothole detection performance of up to 94.78 %.
References
Bučko, Boris, Eva Lieskovská, Katarína Zábovská, and Michal Zábovský. “Computer vision based pothole detection under challenging conditions.” Sensors, 2022, 22, no. 22, 8878.
Ye, Guanting, Jinsheng Qu, Jintai Tao, Wei Dai, Yifei Mao, and Qiang Jin. “Autonomous surface crack identification of concrete structures based on the YOLOv7 algorithm.” Journal of Building Engineering, 2023, 73, 106688.
R. Girshick, J. Donahue, T. Darrell, and J. Malik, “Rich feature hier-archies for accurate object detection and semantic segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2014, pp. 580–587.
R. Girshick, “Fast r-cnn,” in The IEEE International Conference on Computer Vision (ICCV), December 2015.
S. Ren, K. He, R. Girshick, and J. Sun, “Faster r-cnn: Towards real-time object detection with region proposal networks,” in Advances in neural information processing systems, 2015, pp. 91–99.
W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y. Fu, and A. C. Berg, “Ssd: Single shot multibox detector,” in European conference on computer vision. Springer, 2016, pp. 21–37.
T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Doll ´ar, “Focal loss for dense object detection,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 2980–2988.
J. Redmon and A. Farhadi, “Yolov3: An incremental improvement,” arXiv preprint arXiv:1804.02767, 2018.
J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You only look once: Unified, real-time object detection,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 779– 788.
J. Redmon and A. Farhadi, “Yolo9000: better, faster, stronger,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 7263–7271.
Wang, Chien-Yao, Alexey Bochkovskiy, and Hong-Yuan Mark Liao. “YOLOv7: Trainable bagof-freebies sets new state-of-the-art for real-time object detectors.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 7464-7475.
[T. Le, D. Jaw, I. Lin, H. Liu and S. Huang, “An efficient hand detection method based on convolutional neural network,” 2018 7th International Symposium on Next Generation Electronics (ISNE), Taipei, 2018, pp. 1-2, doi: 10.1109/ISNE.2018.8394651
T. Le, S. Huang and D. Jaw, “Cross-Resolution Feature Fusion for Fast Hand Detection in Intelligent Homecare Systems,” in IEEE Sensors Journal, 15 June15, 2019, vol. 19, no. 12, pp. 4696-4704, doi: 10.1109/JSEN.2019.2901259.
Q. -V. Hoang, T. -H. Le and S. -C. Huang, “An Improvement of RetinaNet for Hand Detection in Intelligent Homecare Systems,”2020 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan), Taoyuan, Taiwan, 2020, pp. 1-2, doi: 10.1109/ICCE-Taiwan49838.2020.9258335.
Hoang, Quoc-Viet, Trung-Hieu Le, and Shih-Chia Huang. “Data augmentation for improving SSD performance in rainy weather conditions.” In 2020 IEEE International Conference on Consumer Electronics-Taiwan (ICCE-Taiwan). IEEE, 2020, pp. 1-2.
Hoang, Quoc-Viet, Trung-Hieu Le, and Shih-Chia Huang. “An improvement of RetinaNet for hand detection in intelligent homecare systems.” In 2020 IEEE International Conference on Consumer Electronics-Taiwan (ICCE-Taiwan). IEEE, 2020, pp. 1-2.
Huang, Shih-Chia, Quoc-Viet Hoang, Trung-Hieu Le, Yan-Tsung Peng, Ching-Chun Huang, Cheng Zhang, Benjamin CM Fung, Kai-Han Cheng, and Sha-Wo Huang. “An advanced noise reduction and edge enhancement algorithm.” Sensors, 2021, 21, no. 16, 5391.
Huang, Shih-Chia, Quoc-Viet Hoang, and Trung-Hieu Le. “SFA-Net: A selective features absorption network for object detection in rainy weather conditions.” IEEE Transactions on Neural Networks and Learning Systems, 2022.
Huang, Shih-Chia, Quoc-Viet Hoang, and Da-Wei Jaw. “Self-adaptive feature transformation networks for object detection in low luminance images.” ACM Transactions on Intelligent Systems and Technology (TIST), 2022, 13, no. 1, pp. 1-11.
Huang, Shih-Chia, Da-Wei Jaw, Quoc-Viet Hoang, and Trung-Hieu Le. “3FL-Net: An Efficient Approach for Improving Performance of Lightweight Detectors in Rainy Weather Conditions.” IEEE Transactions on Intelligent Transportation Systems, 2023, 24, no. 4, pp. 4293-4305.
Arya, Deeksha, Hiroya Maeda, Sanjay Kumar Ghosh, Durga Toshniwal, and Yoshihide Sekimoto. “RDD2020: An annotated image dataset for automatic road damage detection using deep learning.” Data in brief, 2021, 36, 107133.
Maeda, Hiroya, Yoshihide Sekimoto, Toshikazu Seto, Takehiro Kashiyama, and Hiroshi Omata. “Road damage detection using deep neural networks with images captured through a smartphone.” arXiv preprint arXiv:1801.09454, 2018.
Tzutalin. Labelimg: A Graphical Image Annotation Tool and Label Object Bounding Boxes in Images. Accessed: Oct. 8, 2020. [Online]. Available: https://github.com/tzutalin/labelImg.
