基于Hough变换的摄像机畸变非量测标定方法

广角镜头在获得大视场的同时也引入了严重的光 学畸变,必须对从广角相机所获得的图像进行畸变校正。本文首先 提取图像中的曲线,依据在单参数除式畸变模型下直线段畸变成像近似呈圆弧的特点,在圆 弧拟合过程中标定出畸变参数与 畸变中心的初始近似值；进而采用一种在Hough变换空间中使 用熵函数估计校正后曲线直线度的方 法,利用熵函数对初始近似值作进一步的优化,最终求取更加精确的畸变参数与畸变中心 。基于工程实例,分别利用圆 弧拟合求初始值方法与本文所提出方法对目标图像进行畸变校正。仿真和真实图像实验表明 ,本文方法仅利用单幅图像即可实 现高精度的标定系数,且方法具有较高的稳定性,操作简单、方便和易于实现。

Non-metric method for camera distortion correction based on Hough transform

Camera with wide angle can enlarge the field of view,but it can al so introduce serious lens distortion at the same time. It is necessary to correct the distorted image obtained from the wide-angle ca mera.First,we extract the distorted lines from the image,and then we use the method based on the property that a straight line can be imaged into a circle,with the single parameter division model.In the process of distorted lines fitting to arcs,the method is u sed to ca librate the initial approximation of the distortion parameter and distortion center.And we propose an entropy function in the Hough space to measure the straightness of the undistorted curves.This entropy function is used to optimize the initial approximation,and then to calculate the accurate final distortion parameters and distortion center.Based on the engineering example,we respe ctively use the arc fitting method and the presented method to correct the target distorted image.The simulated and the experimental results show that the proposed method in this paper can use a single image to obtain high accuracy distortio n calibration coefficient.At the same time,this method has high stability.