基于二次曲线拟合的图像亚像素边缘定位算法  

Sub Pixel Edge Detection Algorithm Based on Guadrativ Curve Fitting

作　　者：赵爱明[1] 

机构地区：[1]哈尔滨理工大学测控技术与通信工程学院,黑龙江哈尔滨150040

出　　处：《哈尔滨理工大学学报》2006年第3期68-70,共3页Journal of Harbin University of Science and Technology

摘　　要：在图像测量系统中,边缘提取精确度直接影响测量精确度.提出一种基于二次曲线拟合方法的图像亚像素边缘提取算法,介绍了基于方形孔径采样定理的边缘定位算法的推导过程.计算机仿真和对实际图像边缘定位的实验结果表明,此亚像素边缘定位算法的定位不确定度优于0.14象素,有效提高了图像边缘提取的定位精确度,从而提高测量精确度.The precision of edge location algorithm is related to the precision of measurement system. To improve the precision of edge location algorithm, this paper offers a method we called sub pixel edge detection algorithm which is based on quadratic curve fitting. And this paper also introduces the developing of edge location algorithm which is based on square aperture sampling. The comparison between computer simulation and actual image edge location experiment results shows that the uncertainty of the sub pixel edge location algorithm is better than 0. 14pixel. It is useful to improve the location precision of image edge location.

关 键 词：二次曲线拟合 边缘提取 亚像素算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] P228[自动化与计算机技术—计算机科学与技术]