Buy article PDF
The purchased file will be sent to you
via email after the payment is completed.
US$ 35
|
Smart Structures and Systems Volume 27, Number 4, April 2021 , pages 593-605 DOI: https://doi.org/10.12989/sss.2021.27.4.593 |
|
|
|
A fast image-stitching algorithm for characterization of cracks in large-scale structures |
||
Linlin Wang, Billie F. Spencer, Jr., Junjie Li and Pan Hu
|
||
| Abstract | ||
| Visual inspection of concrete cracks has been widely used in structural health monitoring (SHM). Capturing highresolution images is an effective method to visualize a complete crack, but it is difficult to show a whole crack from a single high-resolution image. One feasible method is using image stitching technique to stitch several images into a complete crack map. However, the current image stitching method is a computationally intensive process. Numerous images are required to cover large-scale structures with sufficient resolution, this can be computationally prohibitive. To address this problem, an improved image stitching method for crack damage evaluation is proposed, which can quickly stitch the crack images without affecting the quality of the stitching or the resulting images. Rather than first stitching the images together and then determining the crack maps, we propose to first develop the crack maps for the individual images and then stitch them together. The proposed method reduces the number of redundant matching points between the original images by combining their characteristics during image stitching, so it can reduce the calculation time without affecting the quality. Also, the results will not be influenced by the image stitching seam, which can reduce the complexity of the algorithm. Several experimental results are provided in this article to demonstrate that the proposed method can reduce the calculation time without affecting the quality of image stitching and have better robustness than the current method in use. | ||
| Key Words | ||
| visual inspection; concrete crack; image stitching; crack properties | ||
| Address | ||
| (1) Linlin Wang, Junjie Li: Faculty of Infrastructure Engineering, Dalian University of Technology, Dalian 116024, China; (2) Linlin Wang, Billie F. Spencer, Jr., Pan Hu: Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA; (3) Pan Hu: College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, 211106, China. | ||