TRAIN_LIN_DENOIS

Training of linear PCA model for image denoising.



 Description:
  The linear PCA model is trained to describe an input
  class of images corrupted by noise. The training data 
  contains images corrupted by noise and corresponding 
  ground truth. The output dimension of the linear PCA
  is tuned by cross-validation. The objective function 
  is a sum of squared differences between ground truth 
  images and reconstructed images. 

 See also
  PCA, PCAREC, LINPROJ.


Source: train_lin_denois.m

About: Statistical Pattern Recognition Toolbox
(C) 1999-2003, Written by Vojtech Franc and Vaclav Hlavac
Czech Technical University Prague
Faculty of Electrical Engineering
Center for Machine Perception

Modifications:
07-jun-2004, VF
05-may-2004, VF
17-mar-2004, VF