Research On Building Structure Seismic Damage Monitoring Technology Based On Visual Measurement

Due to its geographical location,China has dozens of large and small earthquakes every year.These earthquakes can cause varying degrees of damage to the structure of the building.When the damage accumulates to a certain level,it will cause the building to collapse.The lateral relative displacement that occurs when a building is vibrating is an important basis for analyzing structural damage.Vision measurement method is the most economical and effective monitoring method,and it is becoming a research hotspot.This subject aims to achieve the measurement of the relative displacement of the building structure in both the post-earthquake and epicenter situations,and to achieve the purpose of monitoring the seismic damage of the building structure.The main work of this thesis is as follows:1.Aiming at the structural damage monitoring of buildings after the earthquake,an algorithm for measuring the displacement of deformation of the buildings after the earthquake based on the RGB camera is proposed.It extract the region of interest(ROI)of the image,and use the proposed ROI-based image enhancement method to enhance the image,and use the phase correlation method to convert the image convolution operation in the time-domain to the frequency-domain multiplication operation to improve the matching speed,and use sub-pixel interpolation to interpolate the whole pixel to 1/15 pixel to improve the accuracy of displacement measurement,and use the millimeter/pixel conversion ratio obtained by camera calibration to convert sub-pixel displacement into actual physical displacement.2.Aiming at the monitoring of the jitter environment of the epicenter,an algorithm for measuring the displacement of deformation of the building target under the environment of the epicenter jitter based on the Dynamic Vision Sensor(DVS)is proposed.DVS tracks moving targets,and overcomes the problem of RGB camera motion blur in a jittery environment,and accumulates the information flow generated by DVS into pictures,extracts ROI of the image and filters out noise,and uses an improved adaptive threshold Canny edge detection algorithm to extract the outline of the target circle,and use the Hough transform to coarsely locate the contour of the target circle,and use the least squares fitting circle method on the basis of the coarse positioning to finely locate the target circle to obtain the sub-pixel-level center position.According to the center position of the target circle at the initial moment and the center position of the target circle at the current moment,the movement distance of the target circle is calculated.3.According to the proposed post-earthquake and epicenter displacement measurement algorithm,the displacement measurement equipment was built,and experiments were carried out on the small shaking table of the Department of Civil Engineering of Tsinghua University and the shaking table of the Institute of Engineering Mechanics of China Earthquake Administration(Harbin).These experiments show that the post-earthquake displacement measurement algorithm has higher displacement accuracy than the Orientation Code Matching(OCM)algorithm and the Edge Enhanced Matching(EEM)algorithm,and the speed is 43.3% faster than OCM and 19.6% faster than EEM.The epicenter displacement measurement algorithm can quickly extract the relative displacement between floors,with higher accuracy,and the data of DVS is smaller than that of RGB cameras,and the algorithm processing speed is faster.