Centered Subset Kernel PCA for Denoising
Kernel PCA has been applied to image processing, even though, it is known to have high computational complexity. We introduce centered Subset KPCA for image denoising problems. Subset KPCA has been proposed for reduction of computational complexity of KPCA, however, it does not consider a pre-centering that is often important for image processing. Indeed, pre-centering of Subset KPCA is not straightforward because Subset KPCA utilizes two sets of samples. We propose an efficient algorithm for pre-centering, and provide an algorithm for pre-image. Experimental results show that our method is comparable with a state-of-the-art image denoising method.
Unable to display preview. Download preview PDF.
- 1.Schölkopf, B., Smola, A.J.: Learning with Kernels; Suppot Vector Machines, Regularization, Optimization, and Beyond. MIT Press, Cambridge (2001)Google Scholar
- 3.Mika, S., Schölkopf, B., Smola, A.J., Müller, K.R., Scholz, M., Rätsch, G.: Kernel PCA and de–noising in feature spaces. In: Kearns, M.S., Solla, S.A., Cohn, D.A. (eds.) Advances in Neural Information Processing Systems 11. MIT Press, Cambridge (1999)Google Scholar
- 5.Washizawa, Y.: Subset kernel principal component analysis. In: Proc. of the IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2009), p. 40 (2009)Google Scholar