An Unsupervised Learning Algorithm for Intelligent Image Analysis

This paper presented a new unsupervised learning method to find a set of templates specific to the objects. Kernel PCA, as an unsupervised learning method, is a nonlinear extension of PCA for finding projections that give useful nonlinear descriptors of the data, which gave the system improved performance with continued use by adjusting the clusters, and by creating a new cluster whenever an unusual shape is presented. The learned templates allowed intelligent search of templates for detection, the realistic initialization of object boundaries for segmentation, and the recognition of particular classed for classification. We briefly focused on how to use and work with the kernel-based algorithm in radial basis function neural networks. Finally, a combined kernel PCA RBF neural network model is proposed to cluster shapes for detection and segmentation in a hidden layer, and unsupervised learning to classify objects

[1]  William Nick Street,et al.  Automatic Feature Mining for Personalized Digital Image Retrieval , 2001, MDM/KDD.

[2]  Karl Pearson F.R.S. LIII. On lines and planes of closest fit to systems of points in space , 1901 .

[3]  Sun-Yuan Kung,et al.  A neural network learning algorithm for adaptive principal component extraction (APEX) , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[4]  Kyoung-Mi Lee,et al.  Intelligent image analysis using adaptive resource-allocating network , 2001, Neural Networks for Signal Processing XI: Proceedings of the 2001 IEEE Signal Processing Society Workshop (IEEE Cat. No.01TH8584).

[5]  Paul Scheunders,et al.  A comparison of clustering algorithms applied to color image quantization , 1997, Pattern Recognit. Lett..

[6]  E. Oja Simplified neuron model as a principal component analyzer , 1982, Journal of mathematical biology.