A METHOD FOR IMAGE CLASSIFICATION BASED ON KERNEL PCA
This paper adopts unsupervised on-line shape learning for image analysis tasks, removing the requirement for a pre-defined set of templates and allowing the system to handle novel objects. This learning approach was chosen for its simplicity and extensibility. The results show that the size and shape features are sufficient for accurate object classification. We briefly focused on how to use and work with the kernel-based algorithm in radial basis function neural networks. Kernel PCA, as an unsupervised learning method, is a nonlinear extension of PCA for finding projections that give useful nonlinear descriptors of the data.
Cluster Kernel PCA RBF neural networks Unsupervised learning
SU YAN JIU-FEN ZHAO JIU-LING ZHAO QING-ZHEN LI
Tsinghua University The Second Artillery Engineering Institute
国际会议
2008 International Conference on Machine Learning and Cybernetics(2008机器学习与控制论国际会议)
昆明
英文
718-722
2008-07-12(万方平台首次上网日期,不代表论文的发表时间)