Displacement Monitoring Method Based On Structural Surface Feature Images

The deformation and displacement of the building structure under load are one of the important indicators reflecting the structural stress characteristics.How to quickly and accurately realize structural displacement monitoring is the core problem to be solved.However,traditional displacement monitoring methods such as accelerometers,GPS,and total stations have difficulties in installation,time consuming,and increased cost,making it difficult to achieve real-time,full-field,and convenient needs for building structural positioning and shape-finding.At present,the applications of digital image technology in structural displacement monitoring are mostly based on manually setting marker points.The layout of artificial marker points is limited by the actual structural form,and the installation of marker points has certain difficulties.Therefore,this paper proposes a displacement monitoring method based on the surface features of the structure itself.The main research work of this paper is carried out in three aspects.Structural component image separation method based on improved mean shift algorithm.By analyzing the color distribution of the component area where the surface features of the structure are located,the position of the target component is preliminarily identified based on the inter-frame difference method.Taking the center of the component area as the starting point of the search,the component area of the target structure surface feature is described based on the mean shift algorithm.Through the Bhattacharyya similarity judgment function,the positions of the target structural component are identified in different frame images,and the image of the target component where the structural surface features are located is separated.And the coarse positioning of the structural surface features under complex background is completed.The effectiveness of the proposed method in the image separation of structural members is verified by the truss structure displacement test.Structure surface feature detection and tracking method based on direction angle coding.Using the surface features of the structure itself,such as bolts and holes,the gradient information of the pixel points in the surface feature area is described by the direction angle coding to realize the description of the structure surface feature model.The target structure features are detected and tracked in different frame images.Through camera calibration,the relevant parameters of coordinate transformation and distortion correction are obtained,and the displacement of the target structure features is calculated.And then the structural displacement monitoring is realized.The effectiveness of the proposed method in the detection and tracking of structural surface features is verified by a truss structure displacement test.Displacement monitoring tests based on structural surface feature images.The vibration displacement monitoring test of the frame structure and the displacement monitoring test of the tied arch bridge are carried out.In the frame structure test,the lateral dynamic displacement of the structure is monitored,and the monitoring results of the image method and the laser sensor are compared.The vertical displacement of the deck is monitored in the tie-bar arch bridge test,and the monitoring results of the image method and the total station are compared.The experimental verification of the practical engineering application effectiveness of the proposed method in structural displacement monitoring is completed.