Translator Disclaimer
Paper
8 February 2019 Improvement of geometric calibration algorithm with collinear constraints
Author Affiliations +
Proceedings Volume 10843, 9th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Sensing and Imaging; 108431C (2019) https://doi.org/10.1117/12.2506578
Event: Ninth International Symposium on Advanced Optical Manufacturing and Testing Technologies (AOMATT2018), 2018, Chengdu, China
Abstract
With digital cameras coming into wide-spread use and the intelligence in application system increasingly growing, 3D reconstruction has become an essential part of the vision system. For the sake of achieving it, geometric camera calibration in the context of three-dimensional machine vision must be performed firstly to determine a set of parameters that describe the mapping between 3-D reference coordinates and 2-D image coordinates. In the typical classic method with high calculation accuracy and strong robustness, however, little attention has been paid to initial values of distortion coefficients, and model constraints that make results global optimal. In this paper, we present an improved algorithm based on the traditional calibration method. First, determine exact homography matrices by RANSAC algorithm to reject more error points, solve the initial values of distortion with distortion model, and then constrain the image coordinates of feature points in line by straight lines. Finally, the whole optimized parameters, the suppression of re-projection errors, and the calibration parameters with higher precision are obtained. It becomes a prerequisite for the realization of image fusion and the wide application of computer vision in the field of 3D reconstruction.
© (2019) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Zhao Guan, Weidong Qiao, Jianfeng Yang, Bin Xue, Baogang Lv, and Nange Wang "Improvement of geometric calibration algorithm with collinear constraints", Proc. SPIE 10843, 9th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Sensing and Imaging, 108431C (8 February 2019); https://doi.org/10.1117/12.2506578
PROCEEDINGS
10 PAGES


SHARE
Advertisement
Advertisement
Back to Top