In situ calibration of camera-refraction interface based on analytical refractive imaging equation

TL;DR

This paper introduces an in situ calibration method using the analytical refractive imaging equation to improve camera calibration accuracy through refractive interfaces, outperforming traditional techniques.

Contribution

It presents a novel calibration approach that jointly estimates camera and interface parameters directly from distorted images using the ARI equation.

Findings

Reduces calibration error to 10% of conventional methods.

Maintains computational efficiency comparable to existing methods.

Enhances 3D reconstruction accuracy through improved calibration.

Abstract

Camera calibration is an essential process in photogrammetry, serving as a crucial link between the 2D image coordinate system and the 3D world coordinate system. However, when observations are conducted through refractive interfaces, the refraction effects at these interfaces render traditional calibration methods ineffective, significantly compromising measurement accuracy. To address this challenge, we propose a novel camera calibration method based on the analytical refractive imaging (ARI) equation. The ARI method facilitates accurate estimation of camera parameters from distorted images and enables in-situ joint calibration of both the camera and the refractive interface. The experimental results indicate that the proposed method reduces the error to only 10% of that produced by conventional ray-tracing (RT) method. Moreover, while maintaining comparable computational accuracy and efficiency, it effectively mitigates the local convergence issues that may arise in the polynomial fitting (PF) approach. Finally, reconstruction experiments further confirm the accuracy of the proposed method. Experimental results demonstrate that the proposed method outperforms existing refractive calibration techniques in terms of accuracy while maintaining high precision in 3D reconstruction tasks.