Study on the feasibility of identifying concrete crack width with images acquired by unmanned aerial vehicle
Zhong Xingu1,2 Peng Xiong1,2 Shen Mingyan1,2
1. School of Civil Engineering, Hunan Science and Technology University, Xiangtan 411201, China;
2. Hunan Key Laboratory of Structural Wind and Vibration Control, Hunan Science and Technology University, Xiangtan 411201, China
Abstract：Aiming at the serious problems incurred by the traditional bridge detection vehicle, including obstructing the traffic, being restricted by bridge type, and high maintenance cost, the research was conducted. The unmanned aerial vehicles (UAV) were used for the imaging platform, and the good imaging quality of unmanned aerial vehicles in a hovering state can be proved by making use of the IMETRUM non-contact measurement instrument, according to the imaging quality requirement of above 0.2mm for identification of crack width. By installing three point laser rangefinder to measure the object distance and derive the angle between the imaging plane and the measured plane, the pixel resolution can be calculated and corrected. The image preprocessing program suitable for UAV imaging is designed, the intelligent extraction training model of crack shape based on support vector machine (SVM) is established, and the calculation method for crack width along normal direction is proposed. With the second bridge of Xiangjiang river in Xiangtan taken as an example, the identification of crack width based on UAV is carried out for the real bridge, and then compared with that of the traditional manual test. It is found that the crack width identified by UAV imaging meets the engineering accuracy requirement, and the unmanned aircraft as the working platform to replace the bridge inspection car or support platform is feasible and promising in identifying the shape and width of cracks on the surface of the structure.
