Class distributions on SOM surfaces for feature extraction and object retrieval

详细信息	查看全文 | 推荐本文 |

• 作者：Laaksonen ; Jorma T. ; Markus Koskela ; J. ; Oja ; Erkki
• 关键词：Self-organizing map ; Feature extraction ; Probability distribution ; Entropy ; Mutual information ; Content-based image retrieval
• 刊名：Neural Networks
• 出版年：2004
• 期刊代码：94_08936080
• 类别：cp
• 出版时间：October - November, 2004
• 卷：17
• 期：8-9
• 页码：1121-1133
• 文件大小：263 K

摘要

A Self-Organizing Map (SOM) is typically trained in unsupervised mode, using a large batch of training data. If the data contain semantically related object groupings or classes, subsets of vectors belonging to such user-defined classes can be mapped on the SOM by finding the best matching unit for each vector in the set. The distribution of the data vectors over the map forms a two-dimensional discrete probability density. Even from the same data, qualitatively different distributions can be obtained by using different feature extraction techniques.