Precision and Efficiency in Dam Crack Inspection: A Lightweight Object Detection Method Based on Joint Distillation for Unmanned Aerial Vehicles (UAVs)

被引:0
|
作者
Dong, Hangcheng [1 ]
Wang, Nan [2 ]
Fu, Dongge [1 ]
Wei, Fupeng [3 ]
Liu, Guodong [1 ]
Liu, Bingguo [1 ]
机构
[1] Harbin Inst Technol, Sch Instrumentat Sci & Engn, Harbin 150001, Peoples R China
[2] Hainan Normal Univ, Sch Informat Sci & Technol, Haikou 571158, Peoples R China
[3] North China Univ Water Resources & Elect Power, Sch Informat Engn, Zhengzhou 450045, Peoples R China
关键词
dam crack inspection; object detection; lightweight model; knowledge distillation;
D O I
10.3390/drones8110692
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
Dams in their natural environment will gradually develop cracks and other forms of damage. If not detected and repaired in time, the structural strength of the dam may be reduced, and it may even collapse. Repairing cracks and defects in dams is very important to ensure their normal operation. Traditional detection methods rely on manual inspection, which consumes a lot of time and labor, while deep learning methods can greatly alleviate this problem. However, previous studies have often focused on how to better detect crack defects, with the corresponding image resolution not being particularly high. In this study, targeting the scenario of real-time detection by drones, we propose an automatic detection method for dam crack targets directly on high-resolution remote sensing images. First, for high-resolution remote sensing images, we designed a sliding window processing method and proposed corresponding methods to eliminate redundant detection frames. Then, we introduced a Gaussian distribution in the loss function to calculate the similarity of predicted frames and incorporated a self-attention mechanism in the spatial pooling module to further enhance the detection performance of crack targets at various scales. Finally, we proposed a pruning-after-distillation scheme, using the compressed model as the student and the pre-compression model as the teacher and proposed a joint distillation method that allows more efficient distillation under this compression relationship between teacher and student models. Ultimately, a high-performance target detection model can be deployed in a more lightweight form for field operations such as UAV patrols. Experimental results show that our method achieves an mAP of 80.4%, with a parameter count of only 0.725 M, providing strong support for future tasks such as UAV field inspections.
引用
收藏
页数:20
相关论文
共 50 条
  • [21] HeMoDU: High-Efficiency Multi-Object Detection Algorithm for Unmanned Aerial Vehicles on Urban Roads
    Shi, Hanyi
    Wang, Ningzhi
    Xu, Xinyao
    Qian, Yue
    Zeng, Lingbin
    Zhu, Yi
    SENSORS, 2024, 24 (13)
  • [22] A joint attitude control method for small unmanned aerial vehicles based on prediction incremental backstepping
    Xie, Ting Ting
    Zheng, Jishi
    Chu, Shu-Chuan
    Journal of Information Hiding and Multimedia Signal Processing, 2016, 7 (02): : 277 - 285
  • [23] Real-Time Object Detection and Tracking for Unmanned Aerial Vehicles Based on Convolutional Neural Networks
    Yang, Shao-Yu
    Cheng, Hsu-Yung
    Yu, Chih-Chang
    ELECTRONICS, 2023, 12 (24)
  • [24] Unmanned aerial vehicles object detection based on image haze removal under sea fog conditions
    Wang Pikun
    Wu Ling
    Qi Jiangxin
    Dai Jiashuai
    IET IMAGE PROCESSING, 2022, 16 (10) : 2709 - 2721
  • [25] An adaptive lightweight small object detection method for incremental few-shot scenarios of unmanned surface vehicles
    Wang, Bo
    Jiang, Peng
    Liu, Zhuoyan
    Li, Yueming
    Cao, Jian
    Li, Ye
    ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 133
  • [26] The Method of Increasing the Detection Range of Unmanned Aerial Vehicles In Multiradar Systems Based on Surveillance Radars
    Lishchenko, Vitaliy
    Khudov, Hennadii
    Tiutiunnyk, Vladyslav
    Kuprii, Volodymyr
    Zots, Fedor
    Misiyuk, Galina
    2019 IEEE 39TH INTERNATIONAL CONFERENCE ON ELECTRONICS AND NANOTECHNOLOGY (ELNANO), 2019, : 559 - 562
  • [27] Modeling the Effect of Vegetation Coverage on Unmanned Aerial Vehicles-Based Object Detection: A Study in the Minefield Environment
    Baur, Jasper
    Dewey, Kyle
    Steinberg, Gabriel
    Nitsche, Frank O.
    REMOTE SENSING, 2024, 16 (12)
  • [28] An Object-Based Hierarchical Method for Change Detection Using Unmanned Aerial Vehicle Images
    Qin, Rongjun
    REMOTE SENSING, 2014, 6 (09) : 7911 - 7932
  • [29] A novel high-precision lightweight concrete bridge crack image recognition method based on knowledge distillation
    Fan, Qian
    Xie, Jun
    Xia, Zhanghua
    Tan, Yunhua
    Zhu, Sanfan
    Wang, Jiayu
    STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2024,
  • [30] A Lightweight Anti-Unmanned Aerial Vehicle Detection Method Based on Improved YOLOv11
    Gao, Yunlong
    Xin, Yibing
    Yang, Huan
    Wang, Yongjuan
    DRONES, 2025, 9 (01)