광학 시뮬레이터를 이용한 어안렌즈 배럴 왜곡 추정 및 기하학적 보정

This paper presents a optical simulator based uncorrected barrel distortion estimation and its correction approach based on field of view (FOV) in a fish-eye lens. Radial distortion is estimated by controlling the environment variables in a optical simulator. Base on the estimated distortion, the proposed barrel distortion correction algorithm using FOV effectively provides a wide-angel image. Experimental result show the proposed algorithm can efficiently estimate and correct barrel distortion in a fish-eye lens without the constraints of external environment such as low exposure, radiation change, divergence of light rays, camera shake, to name a few.