摘要
在逆向工程中,为了解决截面线上弱化特征点的提取问题,提出一种基于二阶离散曲率的特征点提取算法。该方法运用了包含高斯核函数曲线的离散曲率表达式来分割曲线,在求解相应的一阶和二阶离散曲率时,选用了不同的尺度因子,由输出曲率的局部极值点来确定截面线的特征点,继而进行特征点的融合。该方法能最大限度地与原有形状特征元保持一致。文中算法与有关文献算法的特征点输出进行比较,实验结果表明所提出算法的适用性和有效性。
To deal with the weak feature extraction of sectional curve in reverse engineering, a feature point extraction method based on two orders discrete curvature is proposed. An expression of discrete curvature with Gaussian nuclear function is used to segment curve. And different scale element is adopted to solve the corresponding one order and two orders discrete curvature, which the feature point of sectional curve is obtained from the local extremum of output curvature. The following work is that some feature points are fused in demand. The propped method guarantees the consistence between feature extraction results and original model feature element. Experimental results show the applicability and efficiency of the proposed method, which compared with the feature point output of relevant literature arithmetic.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2009年第14期114-116,121,共4页
Journal of Wuhan University of Technology
基金
山东省自然科学基金(Y2006F12)
关键词
特征提取
离散曲率
高斯核函数
尺度因子
点融舍
feature extraction
discrete curvature
Gaussian nuclear function
scale element
point fusion
