由于光学镜头会产生几何畸变,并且摄像机成象平面与景物平面间存在着倾角和转角,因而光学镜头摄像机获取的图象存在着非线性的几何畸变,在对图象进行检测和定量分析之前,应消除这些畸变.为此,应用多项式变形技术和双线性插值方法,研究了一种摄像机拍摄图象几何畸变的修正方法,并讨论了控制点的选取方法.实验结果表明,应用三次多项式变形技术,选取少量控制点即可得到理想的校正结果.该方法简便实用,可用于固定条件下摄像机操作平台和生产线工业检测不失真图象的获取任务.实际应用表明,该方法是可靠有效的.

Images with severe geometrical distortions,which are induced by camera lens and position, must be corrected before extracting quantitative measurements.In this paper,we present a geometric calibration method for lens camera by applying polynomial warping and bilinear interpolation,and discusses the selection pattern of control points.First,we implement a general polynomial warping algorithm and a bilinear interpolation algorithm in Visual C++5.0,then we demonstrate that,for common optical lens,images with geometric distortions can be rectified with cubic polynomial warping method using only a few control points.This method is simple and practical,and can be used in camera-fixed situations,such as operating table and on-line industry inspection,to acquire lossless images. Practical application in lottery printing quality inspection shows that the method is reliable and valid.