一种基于现场定标的光电图像畸变校正算法

针对CCD摄像机所采集的图像存在畸变问题,提出了一种基于现场定标法的多项式变形技术和三元卷积算法,对其进行校正.灰阶Sobel算子通过引入衰减因子对图像进行边缘检测得到不失真的灰阶边缘图,然后将灰阶边缘图进行三次样条插值处理,使标准定标图案边缘的定位达到亚像素级,提高了图像边缘检测的精度.利用建立的校正畸变的函数关系式,实现空间几何坐标变换,既保证特殊环境下的图像测量具有足够高的精度,又使光电图像几何畸变校正达到很好的效果.

A Rectification Algorithm for Photoelectric Image with Distortion Based on Method of Onsite Ascertained Standard

A rectification algorithm for the distortional image which are captured with CCD camera is proposed,which introduces the multinomial transformation technique based on the method of onsite ascertained standard and ternary convolution into image rectification. The edges of an image is detected by gray Sobel operator with a attenuation factor. A gray edge image is obtained and is not distortional. Then the gray edge image can be processed using cubic spline interpolation. The location precision of the standard pattern is improved to sub-pixel. It improves the precision of edge detection. Using the established formulas of the function relation for rectification distortion, the spatial geometric coordinate transformation is achieved. It not only ensures the measured image in the special circumstance to be the high precision, but also makes the geometric distortion rectification of photoelectric image achieve good results.