一种扩展小孔成像模型的鱼眼相机矫正与标定方法

A Calibration Method for Fish-eye Cameras Based on Pinhole Model

鱼眼相机由于其超宽的视场范围(Field of view,FOV)(可以达到180°以上),得到越来越广泛的应用.常规的基于小孔成像模型的相机矫正与标定算法在超宽视场的鱼眼成像系统中已经不太适用,为了兼顾小孔模型的特点,本文提出了一种扩展小孔成像模型的鱼眼相机矫正与标定方法.此方法是对小孔成像模型的进一步拓展,不仅具备小孔模型实现简单、适合人眼视觉效果以及相机标定方便等优点,同时将小孔成像模型适用的视场范围扩展到超宽视场领域.其基本思路是:在利用小孔成像模型对鱼眼相机90°左右视场范围进行矫正与标定的基础上,使用非等间距的点阵模板,并结合直线拟合以及自然邻点插值算法,扩展小孔模型适用的视场范围.本文使用鱼眼相机从不同的角度拍摄多幅模板图,完成鱼眼相机的矫正与标定.通过求取的小孔成像模型参数实现相机的标定;对鱼眼相机拍摄的实际场景图进行畸变矫正测试,结果表明此方法能够很好地矫正鱼眼相机存在的畸变,得到符合人眼视觉效果的矫正图;单幅矫正图视场范围达到130°,结合不同角度拍摄的多幅模板图,可把矫正的视场范围扩展到180°.

Fish-eye cameras are widely used in many occasions such as machine vision and so on due to their ultra-wide field of view （FOV） （about 180°）. In this paper, a calibration method based on pinhole model is presented to correct the distortion of the fish-eye camera. First, the pinhole model is used to calibrate the 90° field of view of the fish-eye camera. Then, the unequal gap template of dot-matrix is used. Besides, the line-fitting and natural neighbor interpretation algorithm axe combined to expand the calibration view of the pinhole model. The actual scene calibration results verify the feasibility and effectiveness of this algorithm. This algorithm which extends the calibration range of the pinhole model can correct the distortion of fish-eye camera well and obtain corrective scene graph that fits human eyes to watch. In addition, the calibration with a single image can result in a 130° FOV, while with multiple images being shot from different angles, the calibration can achieve the 180° FOV.