基于直觉模糊距离的图像配准算法  

Image registration algorithm based on intuitionistic fuzzy distance

作　　者：鲁珊[1] 雷英杰[1] 孔韦韦[1] 雷阳[1] 

机构地区：[1]空军工程大学导弹学院,陕西三原713800

出　　处：《控制与决策》2011年第11期1670-1674,共5页Control and Decision

基　　金：国家自然科学基金项目(60773209)

摘　　要：提出一种基于直觉模糊距离的特征匹配算法.首先,从基准图像和待配准图像中提取特征,并将其直觉模糊化;然后,定义特征间的直觉模糊度量方法,从相似程度和不相似程度2个方面对直觉模糊集间距离进行计算;最后,利用直觉模糊集的排序方法解决特征匹配问题,通过构建匹配矩阵实现特征的匹配.实验表明,所提出的基于直觉模糊集的图像配准方法是有效的.A feature matching strategy based on intuitionistic fuzzy distance is developed.Firstly,the feature extracted from the reference and sense image is turned to intuitionistic fuzzy sets.And the method of IFS measurement is defined,which includes the similar measurement and the dissimilar one.Then the problem of feature matching is solved by using the taxis of intuitionistic fuzzy sets.Finally,the matching matrix is constructed to realize the feature matching.Experiments show the good performance of image registration based on intuitionistic fuzzy sets.

关 键 词：图像配准 直觉模糊 特征匹配 相似性测度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]