Fourier-transform-based two-stage camera calibration method with simple periodical pattern

Xiangcheng Chen,Ruimei Fan,Jun Wu,Xiaokai Song,Qing Liu,Yuwei Wang,Yajun Wang,Bo Tao
DOI: https://doi.org/10.1016/j.optlaseng.2020.106121
IF: 5.666
2020-10-01
Optics and Lasers in Engineering
Abstract:<p>Clear and focused pattern images are essential prerequisites for accurate feature detection in traditional camera calibration methods, which introduce numerous limitations in various areas, such as long-distance photogrammetry. A feature detection method robust against defocusing is proposed for extracting the centers or corners of a planar square periodic target. A Fourier transform is employed to calculate two wrapped phase maps from the periodic target images, which are then used to accurately extract the feature points. The calibration procedure is divided into two stages to obtain more accurate results. A rough calibration is performed to calculate the rotation angles between the target and the camera. If the tilt angle is larger than 12°, the corresponding images are removed. Subsequently, the remaining images are used for precise calibration. The simulations and the experiments demonstrate that the proposed method can accurately calibrate a camera with a planar square periodic pattern, even in the case of severe defocusing.</p>
optics
What problem does this paper attempt to address?