计算技术与自动化2023，Vol.42Issue(4)：47-52,6.DOI:10.16339/j.cnki.jsjsyzdh.202304008
基于视觉数据融合和机器学习算法的在役桥梁病害智能检测方法
Intelligent Detection Method of Bridge Diseases in Service Based on Visual Data Fusion and Machine Learning Algorithm
摘要
Abstract
In order to accurately detect the existing bridge diseases,an intelligent detection method of existing bridge diseases based on visual data fusion and machine learning algorithm is studied.UAV aerial photography technology of radar and visual data fusion is adopted to collect and locate the high-definition image of the bridge in service.An in-service bridge disease detection model based on Mask R-CNN convolution neural network is built to train the detection model dynamically through the migration learning method to obtain the optimal in-service bridge disease detection model.To de-noise and en-hance the located high-definition image of the bridge in service,input the detection model for classification and recognition,and output the disease detection results.The experimental results show that this method can quickly and accurately detect a variety of bridge diseases in service,reduce false detection and missed detection,and realize the fine detection of multiple dis-eases in one map.关键词
视觉数据融合/机器学习/在役桥梁/病害检测/无人机巡检/卷积神经网络Key words
visual data fusion/machine learning/bridge in service/disease detection/UAV patrol inspection/convolu-tion neural network分类
交通工程引用本文复制引用
赵琳..基于视觉数据融合和机器学习算法的在役桥梁病害智能检测方法[J].计算技术与自动化,2023,42(4):47-52,6.基金项目
2020年上海市科委优秀技术带头人计划项目(20XD1432400) （20XD1432400）
