| 基于空间矩的灰度边缘亚像元度量精度分析 |
| |
| 引用本文： | 孔丹,李介谷.基于空间矩的灰度边缘亚像元度量精度分析[J].红外与激光工程,1998,27(2):6-10. |
| |
| 作者姓名： | 孔丹 李介谷 |
| |
| 作者单位： | 上海交通大学图像所(孔丹)，中科院自动化所模式识别国家重点实验室(李介谷) |
| |
| 摘 要： | 灰度图像中边缘定位受采样密度限制，亚像元精度度量方法在需要高精度的应用中受到重视，要想知识到底需要多高的分辨率才能达到定精度指标，就有必要分析特征定位方法的最坏情况的精度，文中用分析和实验的方法给出了基于空间矩的边缘定位方法，对于边缘点，边缘直线以及直边缘区域的最坏情况定位精度分析。分析和讨论的结果不但可以用来确定已知形状位置估计的最大误差，而且还可用于对准的形状几何参数，以获得较高的精度。
|
| 关 键 词： | 边缘检测 空间矩 形状定位 图像处理 机器视觉 |
PRECISION ANALYSIS OF SUBPIXEL MEASUREMENT OF EDGES BASED SPATIAL MOMENT IN A GREYLEVEL IMAGE |
| |
| Kong Dan.PRECISION ANALYSIS OF SUBPIXEL MEASUREMENT OF EDGES BASED SPATIAL MOMENT IN A GREYLEVEL IMAGE[J].Infrared and Laser Engineering,1998,27(2):6-10. |
| |
| Authors: | Kong Dan |
| |
| Abstract: | The location precision of an edge measured in a greylevel image is limited by the sam- pling resolution . For those cases, subpixel measurements are at a premium in higher precision applica- tion. It' s useful to know how high a resolution is required for a specified spatial precision. The worst- case precision error ought to be determined . The analysis and experiment results of worst -case preci- sion error for a point, an edge, and a straight-edged shape are proposed in this paper. The conclusion can be used not only to quantify worst-case precision error of measurement of location of known shapes, but also to determine the geometric parameters, so that its shape minimizes the worst-case pre- cision error. |
| |
| Keywords: | Edge detection Spatial moment Shape locating Registration |
| 本文献已被 CNKI 维普 等数据库收录！ |
