Studies on Scene Matching Algorithm of Fuzzy Similarity Metric
-
摘要: 研究了求取图像模糊特征域的计算方法,提出了采用图像模糊集相似度进行景像匹配的方法,并探讨了综合利用多种模糊集相似度算法进行匹配的方法,进一步提出了模糊综合相似度(FSSM)匹配算法.仿真验证表明,FSSM算法具有适应性好、匹配率高等优点,且实时性也较好,为飞行器实时景像匹配提供了一种可行方法.Abstract: Calculations of fuzzy image feature field were studied, the scene matching algorithm adopting similarity metrics of image fuzzy sets was proposed, and the matching methods of combining multiple similarity metrics of fuzzy sets were explored. Further more, the scene-matching algorithm of the fuzzy synthetic similarities (FSSM) was proposed. The simulation results show that the FSSM algorithm has advantages of excellent adaptability, high matching probability, and good real-time ability. As an excellent algorithm of scene matching, it can be used for the real-time positioning of aircrafts.
-
Key words:
- fuzzy sets /
- fuzzy nearness /
- similarity /
- matching
-
[1] 田金文,柳 健,张天序.基于模糊信息理论的景像匹配方法[J].红外与激光工程,1998, 27(3):28~31. [2]何新贵.模糊知识处理的理论与技术[M].第2版.北京:国防工业出版社,1998. [3]Balslev I. Noise tolerance of moment invariants in pattern recognition . Pattern Recognition Letters, 1998,19:1183~1189. [4]汪培庄.模糊集合论及其应用[M].上海:上海科学技术出版社,1983. [5]Chaudhuri B, Rosenfeld A. A modified Hausdorff distance between fuzzy sets[J]. Information Sciences, 1999,118:159~171. [6]周志强.四组合导航系统中的景像匹配技术研究 .北京:北京航空航天大学自动控制系, 1998. [7]Jolion J M, Rosenfeld A. A pyramid framework for early vision[M]. Boston:Kluwer Academic Publishers, 1994.
点击查看大图
计量
- 文章访问数: 3215
- HTML全文浏览量: 99
- PDF下载量: 1379
- 被引次数: 0