A calibration method for defocused cameras based on defocus blur estimation

• Published in: Measurement : journal of the International Measurement Confederation 2024-08, Vol.235, p.115045, Article 115045
• Main Authors: Wan, Jicheng, Zhang, Xuhui, Yang, Wenjuan, Zhang, Chao, Lei, Mengyu, Dong, Zheng
• Format: Article
• Language: English
• Subjects: Camera calibration; Defocus; Defocus blur estimation; Nonlinear optimization; Small planar target
• ISSN: 0263-2241
• DOI: 10.1016/j.measurement.2024.115045

•A defocus camera calibration method based on defocus blur estimation is proposed.•Multi-scale defocus blur estimation is used to obtain the amount of image blur and restore image.•A parameter adaptive particle swarm optimization algorithm (PAPSO) is proposed to optimize the calibration results.•Experiments show that it has good adaptability, high precision and strong operability. The camera calibration is very difficult because of the constraints of space structure and environment in the complex coal mine tunnel excavation construction site. This article proposes a calibration method based on defocus blur estimation for defocused cameras. Firstly, according to the principle of defocus blur, the multi-scale re-blurring method is adopted, and the defocus blur amount of the calibration image is estimated by the gradient ratio of the original image and the re-blurred image. Then, the calibration image is restored by the non-blind deconvolution method, and the camera parameter calibration is completed. A parameter adaptive particle swarm optimization algorithm (PAPSO) is further proposed to optimize the camera parameters. Experimental results demonstrate that the proposed method has significant advantages in terms of flexibility, robustness, and accuracy, and the maximum average reprojection error is 0.237 pixels. It can be applied to the calibration of defocusing camera in coal mine roadway excavation construction site.