A physical based underwater imaging model and camera calibration method

In order to conduct measurement or 3d surface reconstruction in underwater environments, we have to place the camera in a waterproof house and take pictures. Due to the medium of water and air having different refractive indexes, the phenomenon of refraction occurs when light propagates in different medium, which led to the light is no longer a straight line before entering the camera, the incident light will not focus on one point. Traditional calibration model in air is no longer applicable in this situation. By analyzing the physical model of light propagation in the water, and according to the imaging model in air, we propose a nonlinear underwater imaging model and camera calibration method. Through the proposed approach, the physical parameters of our model that affecting the underwater imaging are calibrated. And the measurement experiments show that we can acquire the identical accuracy as in the air, or even better.