摘要
为了实现图像边缘的亚像素定位,针对阶跃形式的边缘类型,提出了一种基于三次样条插值的亚像素边缘检测算法,根据样条插值原理,获得一维边缘的连续灰度分布,通过计算边缘点两侧三次样条函数的二阶导数为零点,实现一维边缘点的亚像素定位;利用一维曲面拟合方法,获得边缘投影方向,根据一维曲面投影方向像素灰度不变理论,利用投影公式将数字窗口中所有三维点投影到相同的投影平面,从而可将二维边缘检测问题转化为一维边缘检测,利用一维边缘检测方法获得边缘点位置,结合投影方向,最终实现二维边缘定位。通过与现有亚像素边缘检测方法的比较,可知算法对噪声具有较好的鲁棒性,同时计算时间相对较快,因此在实际应用中具有较好的适用性。
For the step edge of type form, put forward a sub-pixel edge detection method based on cubic spline interpolation to achieve the image edge sub-pixel localization, according to principle of spline interpolation get continuous gray level distribution of one dimen- sion, then realize the position of one dimension. Using one dimensional curved surface fitting method to obtain the edge projection di- rection , on the basis of one dimension surface projection dil^ction pixel gray-scale invariant theory, then projecting all three dimension points which is in digital window to the same projection plane, so convert two-dimensional edge detection problem into one-dimension. Get edge point location and combine with the projection direction to realize the two-dimensional edge positioning. It is show that the algorithm has good robustness for noise and computation time is relatively fast which compared with the existing sub-pixel edge detection, so has good applicability in practical application.
出处
《控制工程》
CSCD
北大核心
2014年第2期290-293,共4页
Control Engineering of China
基金
吉林省教育厅"十二五"资助项目(吉教科合字[2011]第98号)
关键词
样条插值
亚像素边缘检测
曲面拟合
cubic spline
sub-pixel edge detection
surface fitting
