摘要
为解决高层建筑玻璃幕墙年久失修、经历突发自然灾害以及常年受到高风压载荷导致幕墙变形且难以精确检测的问题,提出一种基于双目立体视觉检测玻璃幕墙变形的方法。该方法首先根据双目成像原理对左右相机进行高精度标定,获取相机的内外参数;然后通过预处理、形态学处理消除图像中的干扰项,并通过霍夫圆变换基于标识点对图像进行特征提取;最后,采用奇异值分解(SVD)和列文伯格-马夸尔特优化算法(LM)计算玻璃幕墙变形前后标识点的三维坐标,从而确定玻璃幕墙的形变。结果表明:借助双目立体视觉测量的方法能够精确检测玻璃幕墙的变形程度,平均相对误差为2.276%,能够有效满足玻璃幕墙变形检测的实际需求。
Aiming at the difficulty in accurate deformation detection for glass curtain wall in high-rise buildings,a new detection method based on binocular stereo vision is proposed to identify its deforma-tion induced by long-term disrepair,sudden natural disasters and perennial high wind pressure loads.The cameras on the left and right sides are first calibrated precisely according to the binocular imaging principle to obtain the internal and external parameters of the cameras.Then the interference terms in the images are eliminated through pre-processing and morphological processing,and the image fea-tures are extracted based on the identification points through the Hough circle transform.Finally,the three-dimensional coordinates of the identification points before and after the deformation of the glass curtain wall are calculated by using singular value decomposition(SVD)and the Levenberg-Mar-quardt(LM)optimization algorithm to determine the deformation of the glass curtain wall.The exper-imental results show that the binocular stereo vision measurement method can accurately detect the de-formation of glass curtain walls with a mean relative error of 2.276%,which can effectively meet the practical needs of glass curtain wall deformation detection.
作者
舒晨洋
王文韫
李寿科
杨景云
SHU Chenyang;WANG Wenyun;LI Shouke;YANG Jingyun(Hunan Provincial Key Laboratory of Health Maintenance for Mechanical Equipment,Hunan University of Science and Technology,Xiangtan 411201,China;College of Civil Engineering,Hunan University,Changsha 410082,China)
出处
《防灾减灾工程学报》
CSCD
北大核心
2024年第1期175-182,共8页
Journal of Disaster Prevention and Mitigation Engineering
基金
重点领域研发计划(2021GK2005)
中央引导地方科技发展资金项目(2022ZYT012)
湖南省自然科学基金项目(2020JJ5187,2021JJ30251)资助。
关键词
玻璃幕墙
双目立体视觉
奇异值分解
LM优化算法
变形检测
glass curtain wall
binocular stereo vision
singular value decomposition
LM optimiza-tion algorithm
deformation detection
