水下相机标定算法研究

Study on underwater camera calibration algorithm

光线在水下成像过程中经过水、玻璃和空气3种折射率不同的介质而发生折射现象。在这种情况下,利用简单针孔成像模型表征水下成像将产生非线性误差。针对这一问题,本文在分析多介质环境中光线的折射效应的基础上建立水下相机模型,并基于该成像模型提出一种水下相机标定算法。在实验室条件下完成水下双目相机的标定工作并与同类标定算法进行比较。实验对比结果显示,基于水下成像模型提出的相机标定算法能够有效的修正折射效应对水下图像产生的畸变影响,更加准确的表征水下成像过程。

In the process of underwater imaging, the light rays refract when the light rays pass through water, glass and air which have different refraction rate in the propagate process. In this case, the refraction leads to a nonlinear error using the simple pinhole camera model characterized the underwater imaging. Aiming at this question, on the basis of analysis the refraction effect, an underwater camera model is built in this paper, then an underwater camera calibration algorithm is proposed. At the same time, an experiment is designed to calibrate the underwater binocular cameras in the laboratory and compare the calibration effect with other calibration algorithm. The results of comparison experiments show that the camera can calibrate properly according to this calibra-tion model, and characterized the underwater imaging process more accurately.