Road crack detection is critical for ensuring road traffic safety, extending the service life of roads, and improving the efficiency of road maintenance management. However, the traditional YOLOv8 model, when applied on mobile devices, faces challenges, such as high network complexity, significant computational resource demands, and slow inference speeds owing to limited computational resources. To address these issues, this paper proposes a model tailored for mobile terminals—Mobile-YOLO. By incorporating the universal inverted bottleneck module and the multi-query attention mechanism, the model significantly reduces network complexity while enhancing computational efficiency for mobile deployment, making it well suited for real-time detection requirements in embedded systems and vehicle-mounted patrol devices. Experimental results showed that Mobile-YOLO improves detection accuracy by 4.1%, mAP50 by 2.76%, and mAP50-95 by 2.56% compared with the baseline YOLOv8, achieving an inference speed of 113 fps, outperforming other lightweight models. Experiments on the NVIDIA Jetson Nano platform further validated its excellent inference performance and low false positive rate, providing an efficient solution for real-world road crack detection in resource-constrained environments.

1. [1] S. Kulkarni, S. Singh, D. Balakrishnan, S. Sharma, S. Devunuri, and S. C. R. Korlapati, “CrackSeg9k: A collection and benchmark for crack segmentation datasets and frameworks,” European Conf. on Computer Vision, pp. 179-195, 2022. https://doi.org/10.1007/978-3-031-25082-8_12
2. [2] P. S. Chakurkar, D. Vora, S. Patil, S. Mishra, and K. Kotecha, “Data-driven approach for AI-based crack detection: Techniques, challenges, and future scope,” Frontiers in Sustainable Cities, Vol.5, Article No.1253627, 2023. https://doi.org/10.3389/frsc.2023.1253627
3. [3] S. B. Li, J. Yang, Z. Wang et al., “Review of development and application of defect detection technology,” Zidonghua Xuebao/Acta Automatica Sinica, Vol.46, No.11, pp. 2319-2336, 2020.
4. [4] M. Kaya and H. S. Bilge, “Deep metric learning: A survey,” Symmetry-Basel, Vol.11, No.9, Article No.1066, 2019. https://doi.org/10.3390/sym11091066
5. [5] D. Reis, J. Kupec, J. Hong, and A. Daoudi, “Real-time flying object detection with YOLOv8,” arXiv:2305.09972, 2023.
6. [6] P. Prasitmeeboon and H. Yau, “Defect detection of particleboards by visual analysis and machine learning,” 2019 5th Int. Conf. on Engineering, Applied Sciences and Technology (ICEAST), 2019. https://doi.org/10.1109/ICEAST.2019.8802526
7. [7] C.-F. Chang, J.-L. Wu, K.-J. Chen, and M.-C. Hsu, “A hybrid defect detection method for compact camera lens,” Adv. Mech. Eng., Vol.9, No.8, Article No.1687814017722949, 2017. https://doi.org/10.1177/1687814017722949
8. [8] F. L. Wang and B. Zuo, “Detection of surface cutting defect on magnet using Fourier image reconstruction,” J. Cent. South Univ., Vol.23, No.5, pp. 1123-1131, 2016. https://doi.org/10.1007/s11771-016-0362-y
9. [9] H. U. Dike, Y. Zhou, K. K. Deveerasetty, and Q. Wu, “Unsupervised learning based on artificial neural network: A review,” Proc. 2018 IEEE Int. Conf. Cyborg Bionic Syst. (CBS), pp. 322-327, 2018. https://doi.org/10.1109/CBS.2018.8612259
10. [10] W. Zaremba, “Recurrent neural network regularization,” arXiv:1409.2329, 2014. https://doi.org/10.48550/arXiv.1409.2329
11. [11] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards real-time object detection with region proposal networks,” IEEE Trans. Pattern Anal. Mach. Intell., Vol.39, No.6, pp. 1137-1149, 2016. https://doi.org/10.1109/TPAMI.2016.2577031
12. [12] K. He, G. Gkioxari, P. Dollár, and R. Girshick, “Mask R-CNN,” Proc. IEEE Int. Conf. Comput. Vis., pp. 2980-2988, 2017. https://doi.org/10.1109/ICCV.2017.322
13. [13] J. Ren and X. Huang, “Defect detection using combined deep autoencoder and classifier for small sample size,” Proc. IEEE 6th Int. Conf. Control Sci. Syst. Eng., pp. 32-35, 2020. https://doi.org/10.1109/ICCSSE50399.2020.9171953
14. [14] C. Sampedro, J. Rodriguez-Vazquez, A. Rodriguez-Ramos, A. Carrio, and P. Campoy, “Deep learning-based system for automatic recognition and diagnosis of electrical insulator strings,” IEEE Access, Vol.7, pp. 101283-101308, 2019. https://doi.org/10.1109/ACCESS.2019.2931144
15. [15] J. Balzategui, L. Eciolaza, and N. Arana-Arexolaleiba, “Defect detection on polycrystalline solar cells using electroluminescence and fully convolutional neural networks,” 2020 IEEE/SICE Int. Symp. on System Integration (SII), pp. 949-953, 2020. https://doi.org/10.1109/SII46433.2020.9026211
16. [16] X. Gao, M. Jian, M. Hu, M. Tanniru, and S. Li, “Faster multi-defect detection system in shield tunnel using combination of FCN and faster RCNN,” Advances in Structural Engineering, Vol.22, No.13, pp. 2907-2921, 2019. https://doi.org/10.1177/1369433219849829
17. [17] W. Zhao, F. Chen, H. Huang, D. Li, and W. Cheng, “A new steel defect detection algorithm based on deep learning,” Computational Intelligence and Neuroscience, Vol.2021, No.1, Article No.5592878, 2021. https://doi.org/10.1155/2021/5592878
18. [18] B. Su, H. Chen, P. Chen, G. Bian, K. Liu, and W. Liu, “Deep learning-based solar-cell manufacturing defect detection with complementary attention network,” IEEE Trans. Ind. Informat., Vol.17, No.6, pp. 4084-4095, 2020. https://doi.org/10.1109/TII.2020.3008021
19. [19] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y. Fu, and A. C. Berg, “SSD: Single shot multibox detector,” Proc. 14th European Conf. Comput. Vis., pp. 21-37, 2016. https://doi.org/10.1007/978-3-319-46448-0_2
20. [20] J. Redmon, “YOLOv3: An incremental improvement,” arXiv:1804.02767, 2018. https://doi.org/10.48550/arXiv.1804.02767
21. [21] C. Zhang, C.-C. Chang, and M. Jamshidi, “Bridge damage detection using a single-stage detector and field inspection images,” arXiv:1812.10590, 2018. https://doi.org/10.48550/arXiv.1812.10590
22. [22] X. Yu, W. Lyu, D. Zhou, C. Wang, and W. Xu, “ES-Net: Efficient scale-aware network for tiny defect detection,” IEEE Trans. Instrum. Meas., Vol.71, pp. 1-14, 2022. https://doi.org/10.1109/TIM.2022.3168897
23. [23] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal speed and accuracy of object detection,” arXiv:2004.10934, 2020. https://doi.org/10.48550/arXiv.2004.10934
24. [24] R. Wang and C. F. Cheung, “CenterNet-based defect detection for additive manufacturing,” Expert Syst. Appl., Vol.188, Article No.116000, 2022. https://doi.org/10.1016/j.eswa.2021.116000
25. [25] A. G. Howard, “MobileNets: Efficient convolutional neural networks for mobile vision applications,” arXiv:1704.04861, 2017. https://doi.org/10.48550/arXiv.1704.04861
26. [26] X. Zhang, X. Zhou, M. Lin, and J. Sun, “ShuffleNet: An extremely efficient convolutional neural network for mobile devices,” Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 6848-6856, 2018. https://doi.org/10.1109/CVPR.2018.00716
27. [27] K. Han, Y. Wang, Q. Tian, J. Guo, C. Xu, and C. Xu, “GhostNet: More features from cheap operations,” Proc. of the IEEE/CVF Conf. on Computer Vision and Pattern Recognition, pp. 1580-1589, 2020. https://doi.org/10.1109/CVPR42600.2020.00165
28. [28] E. J. Jeong, J. Kim, and S. Ha, “TensorRT-based framework and optimization methodology for deep learning inference on Jetson boards,” ACM Trans. on Embedded Computing Systems (TECS), Vol.21, No.5, Article No.51, 2022. https://doi.org/10.1145/3508391
29. [29] J. Mao et al., “MoDNN: Local distributed mobile computing system for deep neural network,” Design, Automation & Test in Europe Conf. & Exhibition (DATE), pp. 1396-1401, 2017. https://doi.org/10.23919/DATE.2017.7927211
30. [30] T. Chen et al., “TVM: An automated End-to-End optimizing compiler for deep learning,” 13th USENIX Symp. on Operating Systems Design and Implementation (OSDI 18), 2018.
31. [31] B. Moons et al., “14.5 envision: A 0.26-to-10TOPS/W subword-parallel dynamic-voltage-accuracy-frequency-scalable convolutional neural network processor in 28nm FDSOI,” 2017 IEEE Int. Solid-State Circuits Conf. (ISSCC), pp. 246-247, 2017.
32. [32] G. Jocher et al., “ultralytics/yolov5: v6. 1-tensorrt, tensorflow edge tpu and openvino export and inference,” Zenodo, 2022.