一种新的基于Dempster-Shafer理论的自适应遥感分类融合方法

被引:2
作者
刘纯平
刘伟强
孔玲
夏德深
机构
[1] 南京理工大学计算机系教研室
[2] 南京理工大学计算机系教研室 南京
[3] 南京
关键词
数据融合; Dempster-Shafer证据理论; 模糊Kohonen神经网络; 遥感; 分类;
D O I
暂无
中图分类号
TP751 [图像处理方法];
学科分类号
081002 ;
摘要
提出了一种基于Dempster-Shafer’s理论和模糊Kohonen神经网络分类融合的方法。该方法融合了非监督神经网络模型和在Dempster-Shafer证据理论框架中使用邻域信息的思想 ,即当一个待识别模式的每个邻域被划分为支持识别框架中某一类的一个证据体时 ,该证据体支持关于该模式隶属关系的某一假设。SPOT遥感数据的分类实验证明 ,该方法同已有的神经网络技术分类方法相比较 ,具有更强的分类能力
引用
收藏
页码:48 / 53
页数:6
相关论文
共 10 条
[1]  
The Research of Data Fusion from Multiple Sources by Evidential Reasoning. FANG Yong. Journal of Remote Sensing . 2000
[2]  
Combining Ambiguous Evidence with Respect to Ambiguous a priori Knowledge, I: Boolean Logic. Mahler R P S. IEEE Transactions on Systems Man and Cybernetics . 1996
[3]  
Dempster - Shafer Theory for Sensor Fusion in Autonomous Mobile Robots. Murphy R R. IEEE Transactions of Robotics and Automation . 1998
[4]  
Pattern recognition.[P]. DIEP THANH A;AVI-ITZHAK HADAR I;GARLAND HARRY T.EP0622750A2,1994-11-02
[5]  
A Mathematical Theory of Evidence. Shafer G. . 1976
[6]  
Combining the results of several neural network classifiers. Rogova G. Neural Networks . 1944
[7]  
Advances in the Dempster-Shafer Theory of Evidence. Yager R R,Fedrizzi M,Kacprzyk J. . 1994
[8]  
Methods of combining multiple classifiers and their applications to handwriting recognition. Xu L,Krzyzak A,Suen C Y. IEEE Trans.Syst. Man. Cybern . 1992
[9]  
The Modified Dempster-Shafer Approach to Classification. Fixsen D,Mahle R P S. IEEE Transactions on Systems Man and Cybernetics . 1997
[10]  
Study of a modified Dempster-Shager approach using an expected utility interval decision rule. Cheaito A,Lecours M. Proceedings of SPIE the International Society for Optical Engineering . 1999