Calibration and Correction for Omnidirectional Image with a Fisheye Lens

Abstract

Omnidirectional vision (omni-vision) with a fisheye lens holds promise of various applications such as global surveillance, robotic navigation and so on. The calibration for the vision system and correction for the distorted image are critical procedure. One method for calibrating the omni-vision system is proposed. The approach involves the utilities of a cylinder on which inner wall there are several straight lines lying. The calibration programs determine the center, radius and gradient of a fisheye lens image. The optimized result is based on the method comparison. Then the calibration parameters are employed for the correction of image distortions. Several imaging rules are conceived for the designs of fish eye lenses. The regulations are discussed respectively and the distortion correction models are generated. An integral distortion correction approach based on these models is developed. The experiments demonstrate the correction algorithm for a fisheye lens image is satisfactory and practicable in real-time process.