摘要
传统建筑外立面和附属设施损伤检测方法主观性强、效率低、对检测人员经验要求较高,难以满足日趋增长的检测需求。为实现建筑外立面和附属设施损伤检测的自动化和智能化,提出了基于无人机自动航线规划功能的图像自动化采集技术和基于YOLOv5x卷积神经网络算法的损伤智能化识别技术,开发了具有损伤智能化识别功能的基于原始图片的损伤检测系统和具有自动出具损伤情况报告功能的基于摄影测量的损伤检测系统。通过工程应用测试了智能化检测系统的损伤检出率与检测效率,对比分析了不同系统的优劣,并提出了建筑外立面和附属设施损伤智能化检测系统进一步升级的方向。
The traditional damage detection methods of building facades and auxiliary facilities are subjective, inefficient and require rich experience of inspectors, which is not able to meet the increasing detection needs. In order to realize the automation and intelligence of damage detection of building facades and auxiliary facilities, this paper puts forward the automatic image acquisition technology based on unmanned aerial vehicle(UAV) automatic route planning function, the intelligent damage identification technology based on YOLOv5x convolution neural network algorithm, and develops the damage detection system based on the original picture with the function of intelligent damage identification and the damage detection system based on photogrammetry with the function of automatically issuing the damage report. The damage detection accuracy and efficiency of the intelligent detection system are tested through a practical engineering, the advantages and disadvantages of the different systems are compared and analyzed, and the direction of further upgrading of the intelligent detection system for building facade and auxiliary facilities is suggested.
作者
陈玲珠
赵宇翔
王卓琳
王易豪
韩文鹤
许清风
CHEN Lingzhu;ZHAO Yuxiang;WANG Zhuolin;WANG Yihao;HAN Wenhe;XU Qingfeng(Shanghai Key Laboratory of Engineering Structure Safety,Shanghai Research Institute of Building Sciences Co.,Ltd.,Shanghai 200032,China;Shanghai Engineering Research Center for Safety Intelligent Control of Building Machinery,Shanghai Research Institute of Building Sciences Co.,Ltd.,Shanghai 200032,China)
出处
《施工技术（中英文）》
CAS
2023年第3期25-30,90,共7页
Construction Technology
基金
国家重点研发计划(2021YFF0602003)。
关键词
外立面
附属设施
无人机
智能识别
损伤检测
facades
auxiliary facilities
unmanned aerial vehicle
intelligent identification
damage detection
