Dimension reduction based on rough set in image mining

Image mining is a nontrivial process to discover valid, novel, potentially useful, and ultimately understandable knowledge from large image sets or image databases. With the rapid development of rough set theory in recent years, more and more people have applied this theory to different research fields. In order to solve the curse of dimensionality in image mining, in this paper, we give our own solution to this problem based on rough set. After introducing the basic concepts of rough set theory and the attribute reduction of information system, we put forward the related algorithms mainly including the partition algorithm and the dimension reduction algorithm. The experimental result has justified the feasibility of the solution based on rough set.

[1]  Ramasamy Uthurusamy,et al.  Data mining and knowledge discovery in databases , 1996, CACM.

[2]  Petra Perner,et al.  Data Mining - Concepts and Techniques , 2002, Künstliche Intell..

[3]  Natalia A. Schmid,et al.  Thresholding method for dimensionality reduction in recognition systems , 2001, IEEE Trans. Inf. Theory.

[4]  Andrzej Skowron,et al.  Rough set rudiments , 1995 .

[5]  David A. Landgrebe,et al.  High dimensional feature reduction via projection pursuit , 1994, Proceedings of IGARSS '94 - 1994 IEEE International Geoscience and Remote Sensing Symposium.

[6]  Domenico Talia,et al.  Eureka! : A Tool for Interactive Knowledge Discovery , 2002, DEXA.

[7]  Padhraic Smyth,et al.  From Data Mining to Knowledge Discovery in Databases , 1996, AI Mag..

[8]  Tsuhan Chen,et al.  Face recognition using mixtures of principal components , 2002, Proceedings. International Conference on Image Processing.

[9]  Mong-Li Lee,et al.  Image Mining: Trends and Developments , 2002, Journal of Intelligent Information Systems.