Research On Holographic Dynamic And Static State Monitoring Method Of Large Bridge Based On Spatiotemporal Image Sequence

The bridge-structure security monitor,on the basis of a series of images,is more informative,convenient and economical than the traditional system(i.e.,the one based on the sensor of measuring points).It is also a good method to take advantage of the development of digital information in the future.This paper systematically studies the holographic structure holography of large bridges morphological feature monitoring and damage identification based on dynamic static rotation photography under the support of the National Natural Science Foundation Project “Research on the method of obtaining the bridge deformation and deducing the structural state based on the differential analysis of image contour(51778094)” and “Evolution and failure process of multiscale analysis method of bridge structure based on the surface shape of spatial-temporal data(51708068)”.The main part of the current research includes:Towards the goal of better monitoring the structure behavior of large bridges,we propose an approach to acquire fixed-axis rotation time-space series images.This work sets the criteria of reference image coordinate system in the frame of the fixed axis rotation space series image.It also analyzes the angle of view and the maximum value of view of the fixed axis rotation,and proposes,based on the structural dynamic and static integration,a novel image acquisition strategy.Besides,current work analyzes the theoretical relationship between the camera rotation and the image overlap degree,and derives the calculation formula of the overlap degree of the adjacent images under the condition of a known rotation angle.A hardware system,which can be characterized as the low cost,adjustable vision range,high precision structure dynamic and static image acquisition,is designed.The three error sources of the system are analyzed,the corresponding error theoretical model is established,and the corresponding strategies are proposed to control the error.According to the criterion of structural deformation continuity,the method to establish the spatial-temporal relationship under static action is proposed.Based on the prior knowledge of bridge structure and spline fitting function,the method of extracting eigen contour lines representing the essential features of structural form is studied,and the eigenfunction of structure form is also constructed.The feature points extraction in machine vision algorithm is constrained by the eigenfunctions of each visual field as boundary conditions,and the tracking algorithm of structural contour lines is formed.At the same time,the dense optical flow is obtained from the eigenfunction.The method is used to constrain the displacement tracking accuracy and improve the calculation efficiency.This method establishes the rotation and acquires the spatial-temporal elevation datasets of the bridge structure by considering time,position and angle as parameters to establish axis.We also study the method of establishing spatial relationship of sequence image according to the perspective relationship and scale invariant transformation of each image horizon;multi-conditions loading experiments were carried out on a self-anchored suspension model bridge.The dynamic and static integrated image monitoring device and dial gauge are used to obtain the deflection change of the model bridge girder,and the accuracy of the device is verified.The advantages of holographic shape curve compared with the finite measurement point data are studied.In order to improve the recognition and tracking accuracy of contour in dynamic image,a framework of image super-resolution algorithm based on edge contour enhancement and photogrammetry is proposed.The algorithm utilizes both the visual measurement and gradient elevation image super-resolution reconstruction algorithm,which is of simple structure and excellent effect.The depth D,shrinkage value v and iteration times M of the tree in the algorithm are determined.We also test the performance of algorithm to describe the edge contour of the structure.In order to verify the effectiveness of the reconstruction algorithm in displacement measurement,an experimental framework is designed to combine image super-resolution algorithm with photogrammetry,and the super-resolution reconstruction algorithm for displacement is studied under different reconstruction ratio and different displacement modes to measure the lifting effect.Combining the shape function of bridge beam eigen contour and dense optical flow algorithm,we propose a method to obtain displacement time history information of eigen contour line at the edge of structure based on segment View Bridge video data under dynamic load.The transformation of the information in time-domain is decoupled to obtain the holographic modal parameters in all directions.The method of visual field mode perspective conversion in the initial state image matrix is studied,and the method of the unity and matching of the mode coordinates of each view area is studied by combining the pixel matching method with the same name,and the method of obtaining the holographic modal parameters of the large bridge structure is formed by the spatial image sequence.Taking 24 m long self-anchored suspension bridge model as the test object,the dynamic parameters of the bridge are verified by the dynamic and static integrated image monitoring device and vibration sensor acquisition model bridge,which verifies the accuracy of the system in the dynamic parameter collection,and studies the characteristics of holographic dynamic parameters.Combining the static holography and deflection curvature damage identification method of the model bridge under different working conditions,the characteristics of the damage identification of the structure are studied,and compared with the data of the conventional displacement mete.Based on the numerical analysis of the simple supported beam,the characteristics of the high-resolution mode shape and curvature mode theory are also studied.The self-anchored suspension model is used to identify the structural damage.The characteristics of damage identification of holographic curvature mode structure are studied in this paper.Compared with the finite measurement point data,the effect of the change rate of different order of curvature mode superposition line as damage identification index is studied.