Author:
Chu Xi,Zhou Zhixiang,Deng Guojun,Duan Xin,Jiang Xin
Abstract
In structural deformation monitoring, traditional methods are mainly based on the deformation data measured at several individual points. As a result, only the discrete deformation, not the overall one, can be obtained, which hinders the researcher from a better and all-round understanding on the structural behavior. At the same time, the surrounding area around the measuring structure is usually complicated, which notably escalates the difficulty in accessing the deformation data. In dealing with the said issues, a digital image-based method is proposed for the overall structural deformation monitoring, utilizing the image perspective transformation and edge detection. Due to the limitation on camera sites, the lens is usually not orthogonal to the measuring structure. As a result, the obtained image cannot be used to extract the deformation data directly. Thus, the perspective transformation algorithm is used to obtain the orthogonal projection image of the test beam under the condition of inclined photography, which enables the direct extraction of deformation data from the original image. Meanwhile, edge detection operators are used to detect the edge of structure’s orthogonal projection image, to further characterize the key feature of structural deformation. Using the operator, the complete deformation data of structural edge are obtained by locating and calibrating the edge pixels. Based on the above, a series of load tests has been carried out using a steel–concrete composite beam to validate the proposed method, with the implementation of traditional dial deformation gauges. It has been found that the extracted edge lines have an obvious sawtooth effect due to the illumination environment. The sawtooth effect makes the extracted edge lines slightly fluctuate around the actual contour of the structure. On this end, the fitting method is applied to minimize the fluctuation and obtain the linear approximation of the actual deflection curve. The deformation data obtained by the proposed method have been compared with the one measured by the dial meters, indicating that the measurement error of the proposed method is less than 5%. However, since the overall deformation data are continuously measured by the proposed method, it can better reflect the overall deformation of the structure, and moreover the structural health state, when compared with the traditional “point” measurements.
Subject
Fluid Flow and Transfer Processes,Computer Science Applications,Process Chemistry and Technology,General Engineering,Instrumentation,General Materials Science
Reference32 articles.
1. Experimental Study on Deflection Monitoring Scheme of Steep Gradient and High Drop Bridge;Liu;J. Highw. Transp. Res. Dev.,2015
2. Study on Application of Close-Range Photogrammetric 3D Reconstruction in Structural Tests;Jiang;Res. Explor. Lab.,2016
3. Probabilistic fatigue assessment of rib-to-deck joints using thickened edge U-ribs;Heng;Steel Compos. Struct.,2020
4. Close-Range Photogrammetry and 3D Imaging, 2nd Edition;Luhmann;Photogramm. Eng. Remote Sens.,2015
5. A comparison between photogrammetry and laser scanning
Cited by
16 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献