Research On The Detection Method Of Displacement Angle Of Structural Specimen Based On Binocular Vision

Through shaking table test,we can study the vibration displacement of the structure.The seismic performance is obtained by analyzing the displacement information.The characteristic points of the test specimen are taken as the specific points of the object,and the displacement,velocity,acceleration,deflection and other motion parameters are obtained from the experiment.The focus of this paper is to detect the displacement of the feature points,through the analysis of the displacement curve,to determine whether the deformation of the structure.This thesis studies the present situation and development trend of visual technology applied to structural displacement detection.Understand the basic theory of displacement detection of shaking table test,including digital image formation principle and its forms and characteristics,the image is used to measure the work involved in the coordinate system,image preprocessing algorithm,target positioning and camera calibration algorithm.A new method of displacement measurement based on binocular stereo vision is proposed,and the key technology of this method is studied in detail.The core contents of this paper include: 1)the calibration plate and the calibration algorithm based on circle array are used to calibrate the system.2)by using the color space transform and Hough transform to detect straight lines,the detection rate of straight line is reduced.According to the Harris corner detection algorithm,the feature points of the image can be obtained,and the number and quality of the corners can be controlled by controlling the threshold value.3)In this thesis,we studied the constraint conditions of feature points matching and the SIFT feature descriptor,and then use the bidirectional matching algorithm based on epipolar constraint.In this thesis,the principle of binocular vision measuring displacement is studied.The method is applied to the displacement detection of structural specimens.The binocular vision system calibration,feature points detection and matching are studied.Firstly,the feasibility of the proposed algorithm is verified by static test,and then the dynamic displacement curve of the structure is obtained by using the algorithm.According to the measurement method proposed in this paper,two kinds of dynamic and static experiments are designed,and the experiment is carried out in the laboratory of shaking table.Firstly,the feasibility and accuracy of the algorithm are verified by static test.The accuracy of the proposed algorithm is verified by static experiments.Then the sine wave excitation experiment is carried out,and the displacement of the characteristic points of the structural specimen is calculated by the algorithm.The displacement curve obtained is basically consistent with the sinusoidal excitation signal.