摘要裂缝是大坝最常见的损伤之一,可反映大坝的受力状态和安全性。针对混凝土坝裂缝传统检测算法速度慢、精度低、泛化性能不足等问题,该文基于目标检测神经网络YOLOX (you only look once x)深度学习目标检测算法,提出一种混凝土坝表观裂缝实时检测方法(YOLOX-dam crack detection,YOLOX-DCD)。该方法对YOLOX目标检测神经网络进行改进,首先在网络结构中加入卷积注意力机制,使网络更关注裂缝特征,提高检测效果;其次引入完全交并比(complete intersection over union,CIoU)作为目标定位损失函数;最后在自制的混凝土坝裂缝数据集上进行实验评估,并与现有的多种目标检测神经网络进行对比。结果表明：该文所提方法具有速度快、精度高、参数少的特点,且明显优于经典目标检测算法。因此,该文所提方法能满足混凝土坝裂缝检测高效、精确、实时的要求,可为混凝土坝裂缝检测提供技术支持。
Abstract：As a major part of water conservancy infrastructure, dams play an important role in economic construction and social development. Cracks are one of the most common types of damage to dams, destroying the overall structure and affecting the durability, strength, and stability of the structure. Therefore, regular and systematic crack detection of concrete dams is of great importance to ensure their safe and stable operation. However, the traditional concrete dam crack detection technology suffers from slow speed, low precision, and insufficient generalization performance, bringing difficulty in meeting the requirements of concrete dam crack detection. Therefore, the objective of this study is to develop an efficient, accurate, and real-time concrete dam crack detection technology. Existing crack detection methods based on semantic segmentation algorithms run slowly and detect concrete cracks in real time with difficulty. In addition, the dam operation environment is harsh, resulting in complex image backgrounds and inconspicuous crack image features, increasing the difficulty of identification. This study proposes a real-time detection method for concrete dam cracks based on deep learning object detection method you only look once x (YOLOX), called YOLOX-dam crack detection (YOLOX-DCD), to address the problems of slow speed, low accuracy, and insufficient generalization of the traditional detection techniques for concrete dam cracks. This method improves the performance of YOLOX to detect concrete dam cracks. First, a lightweight convolutional block attention module (CBAM) is added to the network structure, which integrates the spatial attention mechanism with the channel attention mechanism. The CBAM makes the network pay more attention to crack features and improves detection performance. Second, a complete intersection over union (CIoU) is introduced to replace IoU as the loss function. The CIoU incorporates the normalized distance between the predicted box and the target box and summarizes three geometric factors in bounding box regression, i.e., overlap area, central point distance, and aspect ratio, thereby improving the convergence speed and detection performance of the algorithm. The experimental evaluation was conducted on a self-made concrete dam crack dataset. Ablation experiments were performed on each improved module, and the results showed that the improved method proposed in this paper effectively improved the detection accuracy of the model and maintained a high detection speed. The proposed model had an AP0.5 on the test set of 90.84% and an F1 of 87.74%, which were higher than those of various existing object detection methods. The FPS of the model was 65, and the detection speed was faster. The model was small, with a size of 25.67 MB, and could be deployed on a mobile terminal for real-time crack detection. In this study, a CBAM and the CIoU loss function are added to the YOLOX network, which make the network pay more attention to crack characteristics and improves the detection performance for concrete dam cracks. Experiments reveal that the method in this paper has fast speed, high precision, and few parameters and is obviously better than the classical object detection algorithms. Therefore, the proposed method meets the requirements of efficient, accurate, and real-time crack detection of concrete dams and is promising for providing a technical means for crack detection.
黄贲, 康飞, 唐玉. 基于目标检测的混凝土坝裂缝实时检测方法[J]. 清华大学学报（自然科学版）, 2023, 63(7): 1078-1086.
HUANG Ben, KANG Fei, TANG Yu. A real-time detection method for concrete dam cracks based on an object detection algorithm. Journal of Tsinghua University(Science and Technology), 2023, 63(7): 1078-1086.
[1] 何金平.大坝安全监测理论与应用[M].北京:中国水利水电出版社, 2010. HE J P. Dam safety monitoring theory and application[M]. Beijing:China Water Conservancy and Hydropower Press, 2010.(in Chinese) [2] 吴中如.重大水工混凝土结构病害检测与健康诊断[M].北京:高等教育出版社, 2005. WU Z R. Safety diagnosis and hidden defects detection of major hydraulic concrete structures[M]. Beijing:Higher Education Press, 2006.(in Chinese) [3] WU Z R, LI J, GU C S, et al. Review on hidden trouble detection and health diagnosis of hydraulic concrete structures[J]. Science in China Series E:Technological Sciences, 2007, 50(1):34-50. [4] 钮新强.大坝安全诊断与加固技术[J].水利学报, 2007(S1):60-64. NIU X Q. Technique for diagnosing and strengthening dam safety[J]. Journal of Hydraulic Engineering, 2007(S1):60-64.(in Chinese) [5] 苏雨.基于机器学习的某混凝土坝裂缝成因与预警研究[D].长沙:长沙理工大学, 2017. SU Y. Crack cause and early warning analysis of a concrete dam based on machine learning[D]. Changsha:Changsha University of Science&Technology, 2017.(in Chinese) [6] 黄朝君,杨小云,夏杰.丹江口初期工程大坝上游面水上裂缝检查与处理[J].人民长江, 2015, 46(6):45-48, 74. HUANG C J, YANG X Y, XIA J. Inspection and treatment of overwater crack of upstream surface of first-stage project of Danjiangkou dam[J]. Yangtze River, 2015, 46(6):45-48, 74.(in Chinese) [7] 梅智.基于无人机图像的混凝土坝表观裂缝监测研究[D].大连:大连理工大学, 2020. MEI Z. Research on the monitoring of concrete dam surface crack based on UAV image[D]. Dalian:Dalian University of Technology, 2020.(in Chinese) [8] 冯春成,张华,汪双,等.水电站溢流坝表观裂缝损伤智能检测方法研究[J].自动化与仪表, 2021, 36(6):55-60. FENG C C, ZHANG H, WANG S, et al. Research on intelligent detection method for crack damage of overflow dam of hydropower station[J]. Automation&Instrumentation, 2021, 36(6):55-60.(in Chinese) [9] 陈荣敏,王皓冉,汪双,等.基于无人机的坝面裂纹缺陷智能检测方法[J].水利水电科技进展, 2021, 41(6):7-12. CHEN R M, WANG H R, WANG S, et al. Intelligent detection method of crack defects on dam surface based on UAV[J]. Advances in Science and Technology of Water Resources, 2021, 41(6):7-12.(in Chinese) [10] 王琳琳,李俊杰,康飞,等.基于无人机图像拼接技术的大坝健康监测方法[J].人民长江, 2021, 52(12):236-240. WANG L L, LI J J, KANG F, et al. Dam health monitoring method based on image mosaic technology of unmanned aearial vehicle[J]. Yangtze River, 2021, 52(12):236-240.(in Chinese) [11] 陈从平,聂葳,吴喆,等.基于视觉机器人的大坝水下表面裂缝检测系统设计[J].三峡大学学报(自然科学版), 2016, 38(5):72-74, 86. CHEN C P, NIE W, WU Z, et al. Design of underwater dam surface crack detection system based on visual robot[J]. Journal of China Three Gorges University (Natural Sciences), 2016, 38(5):72-74, 86.(in Chinese) [12] LI Y T, BAO T F, HUANG X J, et al. Underwater crack pixel-wise identification and quantification for dams via lightweight semantic segmentation and transfer learning[J]. Automation in Construction, 2022, 144:104600. [13] FENG C C, ZHANG H, WANG H R, et al. Automatic pixel-level crack detection on dam surface using deep convolutional network[J]. Sensors, 2020, 20(7):2069. [14] DUNG C V, ANH L D. Autonomous concrete crack detection using deep fully convolutional neural network[J]. Automation in Construction, 2019, 99:52-58. [15] 陈波,张华,汪双,等.基于全卷积神经网络的坝面裂纹检测方法研究[J].水力发电学报, 2020, 39(7):52-60. CHEN B, ZHANG H, WANG S, et al. Study on detection method of dam surface cracks based on full convolution neural network[J]. Journal of Hydroelectric Engineering, 2020, 39(7):52-60.(in Chinese) [16] 任秋兵,李明超,沈扬,等.水工混凝土裂缝像素级形态分割与特征量化方法[J].水力发电学报, 2021, 40(2):234-246. REN Q B, LI M C, SHEN Y, et al. Pixel-level shape segmentation and feature quantification of hydraulic concrete cracks based on digital images[J]. Journal of Hydroelectric Engineering, 2021, 40(2):234-246.(in Chinese) [17] LIU W, ANGUELOV D, ERHAN D, et al. SSD:Single shot multibox detector[C]//14th European Conference on Computer Vision. Amsterdam, The Netherlands:Springer, 2016:21-37. [18] REDOMN J, DIVVALA S, GIRSHICK R, et al. You only look once:Unified, real-time object detection[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA:IEEE, 2016:779-788. [19] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision. Santiago, Chile:IEEE, 2015:1440-1448. [20] CHA Y J, CHOI W, SUH G, et al. Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types[J]. Computer-Aided Civil and Infrastructure Engineering, 2018, 33(9):731-747. [21] WANG N N, ZHAO X F, ZHAO P, et al. Automatic damage detection of historic masonry buildings based on mobile deep learning[J]. Automation in Construction, 2019, 103:53-66. [22] XU Y, WEI S Y, BAO Y Q, et al. Automatic seismic damage identification of reinforced concrete columns from images by a region-based deep convolutional neural network[J]. Structural Control and Health Monitoring, 2019, 26(3):e2313. [23] 章世祥,张汉成,李西芝,等.基于机器视觉的路面裂缝病害多目标识别研究[J].公路交通科技, 2021, 38(3):30-39. ZHANG S X, ZHANG H C, LI X Z, et al. Study on multi-objective identification of pavement cracks based on machine vision[J]. Journal of Highway and Transportation Research and Development, 2021, 38(3):30-39.(in Chinese) [24] GE Z, LIU S T, WANG F, et al. YOLOx:Exceeding yolo series in 2021[J/OL]. arXiv.(2021-08-06)[2022-10-31]. https://arxiv.org/abs/2107.08430. [25] WOO S, PARK J, LEE J Y, et al. CBAM:Convolutional block attention module[C]//Proceedings of the 15th European Conference on Computer Vision. Munich, Germany:Springer, 2018:3-19. [26] ZHENG Z H, WANG P, LIU W, et al. Distance-IoU loss:Faster and better learning for bounding box regression[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(7):12993-13000.[W
2023, Vol. 63 
