Affiliation:
1. School of Civil Engineering, Qingdao University of Technology, Qingdao 266033, China
Abstract
The crack detection method based on image processing has been a new achievement in the field of civil engineering inspection in recent years. Column piers are generally used in bridge structures. When a digital camera collects cracks on the pier surface, the loss of crack dimension information leads to errors in crack detection results. In this paper, an image stitching method based on Speed-Up Robust Features (SURFs) is adopted to stitch the surface crack images captured from different angles into a complete crack image to improve the accuracy of the crack detection method based on image processing in curved structures. Based on the proposed method, simulated crack tests of vertical, inclined, and transverse cracks on five different structural surfaces were conducted. The results showed that the influence of structural curvature on the measurement results of vertical cracks is very small and can be ignored. Nevertheless, the loss of depth information at both ends of curved cracks will lead to errors in crack measurement outcomes, and the factors that affect the precision of crack detection include the curvature of the surface and the length of the crack. Compared with inclined cracks, the structural curvature significantly influences the measurement results of transverse cracks, especially the length measurement results of transverse cracks. The image stitching method can effectively reduce the errors caused by the structural curved surface, and the stitching effect of three images is better than that of two images.
Funder
National Key Research and Development Program
Key Research and Development Program of Shandong Province
Reference20 articles.
1. Automatic multi-image stitching for concrete bridge inspection by combining point and line features;Xie;Autom. Constr.,2018
2. Condition monitoring of bridges with non-contact testing technologies;Dabous;Autom. Constr.,2020
3. Multi-image stitching and scene reconstruction for evaluating defect evolution in structures;Jahanshahi;Struct. Health Monit.,2011
4. Fujita, Y., Mitani, Y., and Hamamoto, Y. (2006, January 20–24). A method for crack detection on a concrete structure. Proceedings of the 18th International Conference on Pattern Recognition (ICPR’06), Hong Kong, China.
5. Image-based crack detection for real concrete surfaces;Yamaguchi;IEEJ Trans. Electr. Electron. Eng.,2008