深度学习的无人机 + 5G公路巡检系统
UAV + 5G Highway Inspection System of Deep Learning

DOI: 10.12677/CSA.2021.116182, PP. 1763-1771

Keywords: 无人机，5G，深度学习，公路巡检
UAV, 5G, Deep Leaning, Highway Inspection

Full-Text   Cite this paper   Add to My Lib

Abstract:

公路固定监控摄像头存在视距范围小，监控死角等问题，难以满足公路监控的需求，而无人机具有灵活性强、效率高、覆盖范围大等优点，且其与5G结合可以在公路监控领域充分发挥其优势。基于该背景，文章对深度学习的无人机 + 5G巡检系统进行了探讨，5G帮助无人机解决传输距离及传输时延的问题，使其在公路巡检成为可能，并且针对无人机高空巡检视频处理其存在的主要问题——检测准确率与跟踪实时性，使用相应深度学习算法进行实验，最后根据实验效果进行分析。
Highway fixed surveillance cameras have some problems such as small visual range and dead corners, which are difficult to meet the needs of highway monitoring. UAV (Unmanned Aerial Vehicle) has the advantages of strong flexibility, high efficiency, and large coverage. The combination of UAV and 5G can play a huge role in the field of highway monitoring. Based on this background, the paper discusses the UAV + 5G inspection system of deep learning. 5G helps UAV solve the problems of transmission distance and transmission delay, making it possible to conduct road inspections. For the main problems of detection accuracy and real-time tracking of video processing in high-altitude inspection, the paper uses the corresponding deep learning algorithm to conduct experiments, and finally analyze according to the experimental results.

References

[1]	黄凯, 王文辉. 利用无人机开展高速公路监控的可行性分析[J]. 公路交通科技(应用技术版), 2020, 16(10), 366-369.
[2]	汪琦. 智慧监控在高速公路信息化中的应用[J]. 四川水泥, 2019(8): 167.
[3]	冯力, 魏一, 教颖辉. 5G时代, 无人机技术在政务空地一体化智能调度系统中的应用[J]. 中国安防, 2020(8): 82-85.
[4]	刘照启, 张蔚然. 植保无人机的优缺点分析与前景展望[J]. 湖北农机化, 2020(9): 6-7.
[5]	张朝阳, 李星宏. 5G在无人驾驶汽车中的应用前景分析[J]. 无人系统技术, 2019, 2(2): 65-68.
[6]	The Times of Israel (2018) Israel Police Roll out Traffic Drones to Catch Reckless Drivers, Toi Staff. https://www.timesofisrael.com/israel-police-roll-out-traffic-drones-to-catch-recklessdrivers/
[7]	Wang, L., Chen, F. and Yin, H. (2016) Detecting and Tracking Vehicles in Traffic by Unmanned Aerial Vehicles. Automation in Con-struction, 72, 294-308. https://doi.org/10.1016/j.autcon.2016.05.008
[8]	田芳, 王珂. 基于5G + 4K的无人机高速公路自动化巡检研究与应用[J]. 江苏科技信息, 2021, 38(6): 55-59.
[9]	https://github.com/ultralytics/yolov5
[10]	Bochkovskiy, A., Wang, C.-Y. and Liao, H.-Y.M. (2020) YOLOv4: Optimal Speed and Accuracy of Object Detection. arXiv:2004.10934.
[11]	Wang, C.-Y., Liao, H.-Y.M., Yeh, I.-H., Wu, Y.-H., Chen, P.-Y. and Hsieh, J.-W. (2019) CSPNet: A New Backbone That Can Enhance Learning Capabil-ity of CNN. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Seattle, WA, 14-19 June 2020, 1571-1580. arXiv:1911.11929 https://doi.org/10.1109/CVPRW50498.2020.00203
[12]	Lin, T.-Y., Dollár, P., Girshick, R., He, K., Hariharan, B. and Belongie, S. (2017) Feature Pyramid Networks for Object Detection. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, 21-26 July 2017, 936-944. arXiv:1612.03144v2 https://doi.org/10.1109/CVPR.2017.106
[13]	Li, H.C., Xiong, P.F., An, J. and Wang, L.X. (2018) Pyramid Atten-tion Network for Semantic Segmentation. arXiv:1805.10180v3
[14]	Wojke, N., Bewley, A. and Paulus, D. (2017) Simple Online and Realtime Tracking with a Deep Association Metric. 2017 IEEE International Conference on Image Processing (ICIP), Beijing, 17-20 September 2017, 3645-3649. arXiv:1703.07402 https://doi.org/10.1109/ICIP.2017.8296962

Full-Text