基于模糊聚类的标识点图形识别方法  

Method of Marked Point Recognition Based on Fuzzy Cluster

作　　者：石磊[1,2] 孙凯明[1,2] 王刚[2] 甄海涛[2] 

机构地区：[1]黑龙江省科学院高技术研究院,哈尔滨150020 [2]黑龙江省科学院自动化研究所,黑龙江哈尔滨15009

出　　处：《自动化技术与应用》2015年第12期90-92,共3页Techniques of Automation and Applications

摘　　要：提出模糊聚类应用于三维测量中标识点的识别方法 ,通过计算标志点间的欧式距离和模糊矩阵系数可以去除具有成像缺陷的标志点图形,根据标志点与标识样品间的欧式距离可以判断出标志点的形状,此方法应用于三维测量可以提高标志点识别能力和质量。Fuzzy cluster is applied to identify marked point in the three-dimensional measurement. Calculating the Euclidean distance between the marked points and fuzzy matrix coefficient can remove the defective marked points. According to a marked point with the Euclidean distance among perfect sample marked points can determine the shape of a marked point. This method is applied to the three-dimensional measurement that can improve marked point recognition ability and quality.

关 键 词：模糊聚类 标志点 识别方法 

分 类 号：TP291.4[自动化与计算机技术—检测技术与自动化装置]