A method for image classification based on Kernel PCA
Yan Ping Su, Jiufen Zhao, Jiu-Ling Zhao, Qingzhen Li · 2008
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.