BRIDGE UNDERWATER STRUCTURAL DEFECTS DETECTION BASED ON FUSION IMAGE ENHANCEMENT AND IMPROVED YOLOV 7

被引：0

作者：

Li Z.-R. ^{[1
]}

Liu A.-R. ^{[1
]}

Chen B.-C. ^{[1
]}

Wang J.-L. ^{[1
]}

Lan T. ^{[2
]}

Wang B.-X. ^{[3
]}

机构：

[1] Research Center of Wind Engineering and Engineering Vibration, Guangdong, Guangzhou

[2] Guangzhou Grand Engineering Inspection and Consulting Co., Ltd., Guangdong, Guangzhou

[3] School of Safety Engineering and Emergency Management, Shijiazhuang Railway University, Hebei, Shijiazhuang

来源：

Gongcheng Lixue/Engineering Mechanics | 2024年 / 41卷

关键词：

attention module; bridge underwater structure detection; defect recognition; image enhancement; underwater vehicle;

D O I：

10.6052/j.issn.1000-4750.2023.05.S046

中图分类号：

学科分类号：

摘要：

This study presents a new method for automatic identification of structural defects by using ROV. First, low-quality fault photos are obtained under muddy water to expand the dataset. Next, the cascaded Water-Net underwater image improvement method is used to create high-quality images from low-quality photos. The polarized self-attention module is employed to preserve the high resolution output of the improved image, thus allowing the image enhancement and target detection to work together efficiently and enhance the detection accuracy, in order to counteract the hindrance caused by the discrepancy between image enhancement and target detection. Given that low-quality image data annotation would necessarily contain low-quality examples, the WioU loss function is used to replace the loss function in the original YOLOv7-tiny network model to improve generalization performance. The experimental results reveal that, compared with the original network, the revised network model retains defect recognition accuracy while retaining high picture contrast, sharper details, and superb visual effects, as well as dramatically improving missed detections and misjudgments. © 2024 Tsinghua University. All rights reserved.

引用

页码：245 / 252

页数：7

共 26 条

[1] LIU Kang, LIU Junli, YU Wencheng, Statistical analysis of bridge collapse caused by floods in 2007 ~ 2015, Urban Roads Bridges & Flood Control, 1, pp. 90-92, (2017)
[2] JI Bohai, FU Zhongqiu, Analysis of Chinese bridge collapse accident causes in recent years, China Civil Engineering Journal, 43, pp. 495-498, (2010)
[3] ZHU Qiankun, CHEN Jianbang, ZHANG Qiong, Et al., A non-contact recognition for deflection influence line of footbridge based on computer vision, Engineering Mechanics, 38, 8, pp. 145-153, (2021)
[4] YANG Na, ZHANG Chong, LI Tianhao, Design of crack monitoring system for Chinese ancient wooden buildings based on UAV and CV, Engineering Mechanics, 38, 3, pp. 27-39, (2021)
[5] HAN Xiaojian, ZHAO Zhicheng, SHEN Zejiang, Detect diseases on the surface of bridge structures by convolution neural network, Structural Engineers, 35, 2, pp. 106-111, (2019)
[6] SI Xinhua, LIU Dayang, ZHANG Peng, Et al., Bridge steel structural apparent defects automatic detection based on YOLOv4, Highway, 67, 8, pp. 399-402, (2022)
[7] RUAN Xiaoli, WANG Bo, WU Jufeng, Et al., Identification of spalled concrete and exposed reinforcement in reinforced concrete bridge based on deep learning, World Bridges, 48, 6, pp. 88-92, (2020)
[8] ZHANG Shixiang, ZHANG Hancheng, LI Xizhi, Et al., Study on multi-objective identification of pavement cracks based on machine vision, Journal of Highway and Transportation Research and Development, 38, 3, pp. 30-39, (2021)
[9] REN Y P, HUANG J S, HONG Z Y, Et al., Image-based concrete crack detection in tunnels using deep fully convolutional networks, Construction and Building Materials, 234, (2020)
[10] HUANG H W, LI Q T, ZHANG D M., Deep learning based image recognition for crack and leakage defects of metro shield tunnel [J], Tunnelling and Underground Space Technology, 77, pp. 166-176, (2018)

← 1 2 3 →