Back to articles
Volume: 4 | Article ID: art00107
NTF vs. PCA Features for Searching in a Spectral Image Database
  DOI :  10.2352/CGIV.2008.4.1.art00107  Published OnlineJanuary 2008

A technique for searching in a spectral image database is proposed in this study. It is based on a similarity measure between spectral image features. New and convenient spectral image features are introduced and compared here. Nonnegative tensor factorization (NTF) and principal component analysis (PCA) are applied in a spectral image domain to characterize colors of a spectral image. A new way of NTF with a multiresolution approach is used to accelerate the time complexity in the extraction of the features.The proposed method is implemented and tested with a spectral image database. The images from the database are ordered according to the similarity between them and the tested image. Three similarity measures were applied in the two spectral image feature spaces. The results of the experiments are visually represented. The best combination of the spectral image feature and similarity measure in our opinion is proposed during a discussion part. Also further work will be proposed.

Subject Areas :
Views 2
Downloads 0
 articleview.views 2
 articleview.downloads 0
  Cite this article 

Alexey Andriyashin, Arto Kaarna, Timo Jaaskelainen, Jussi Parkkinen, "NTF vs. PCA Features for Searching in a Spectral Image Databasein Proc. IS&T CGIV 2008/MCS'08 4th European Conf. on Colour in Graphics, Imaging, and Vision 10th Int'l Symp. on Multispectral Colour Science,  2008,  pp 499 - 504,

 Copy citation
  Copyright statement 
Copyright © Society for Imaging Science and Technology 2008
Conference on Colour in Graphics, Imaging, and Vision
conf colour graph imag vis
Society of Imaging Science and Technology
7003 Kilworth Lane, Springfield, VA 22151, USA