Abstract
The Gaussian and Gaussian-sigmoidal primaries are introduced as initialization functions in the nonnegative matrix factorization (NNMF) method to achieve more consistent error statistics. The proposed NNMF method with multiplicative update algorithm is employed for reconstruction of spectral data of Munsell and Colorchecker DC datasets from the corresponding CIEXYZ tristimulus values and the resultant achievements are compared with those obtained from PCA and the classical NNMF techniques. Both colorimetric and spectrophotometric errors between the original and the reconstructed spectra are used for the evaluation of reconstruction methods. While run to run variability of the classical NNMF method with multiplicative update algorithm is significantly remarkable, the stability of the suggested technique is fairly high, similar to the PCA method. The ensuing findings show that the embedding of Gaussian initializing vectors in NNMF method provide the all positive bases that enhance both the accuracy of Gaussian method and the reliability of NNMF technique.
Similar content being viewed by others
References
C. J. Hawkyard: J. Soc. Dye Col. 109 (1993) 246.
N. Eslahi, S. H. Amirshahi, and F. Agahian: Opt. Rev. 16 (2009) 296.
H. S. Fairman and M. H. Brill: Color Res. Appl. 29 (2004) 104.
C. J. Hawkyard: J. Soc. Dye Col. 110 (1994) 386.
Y. Sun, F. D. Fracchia, T. W. Calvert, and M. S. Drew: IEEE Comput. Graph. 19 (1999) 61.
K. Ansari, S. H. Amirashahi, and S. Moradian: Color Technol. 122 (2006) 128.
D. Dupont: Color Res. Appl. 27 (2002) 88.
C. J. Hawkyard: J. Soc. Dye Col. 109 (1993) 323.
G. Wang, C. Li, and M. R. Luo: Color Res. Appl. 30 (2005) 283.
J. B. Cohen: Psychon 1 (1964) 369.
A. García-Beltrán, J. L. Nieves, J. Hernández-Andrés, and J. Romero: Color Res. Appl. 23 (1998) 39.
L. T. Maloney: J. Opt. Soc. Am. A 3 (1986) 1673.
F. Agahian, S. A. Amirshah, and S. H. Amirshahi: Color Res. Appl. 33 (2008) 360.
D. Tzeng and R. S. Berns: Color Res. Appl. 30 (2005) 84.
D. D. Lee and H. S. Seung: Nature 401 (1999) 788.
R. S. Berns: Color Res. Appl. 22 (1997) 318.
G. Buchsbaum and O. Bloch: Vision Res. 42 (2002) 559.
S. H. Amirshahi and S. A. Amirshahi: Opt. Rev. 17 (2010) 562.
N. Attarchi and S. H. Amirshahi: Color Res. Appl. 34 (2009) 26.
E. M. Valero, J. L. Nieves, S. M. C. Nascimento, K. Amano, and D. H. Foster: Color Res. Appl. 32 (2007) 352.
S. Westland and C. Ripamonti: C. Computational Color Science Using MATLAB (Wiley, Chichester, U.K., 2004).
A. Cichochi, R. Zdunek, A. H. Phan, and S. Amari: Nonnegative Matrix and Tensor Factorizations: Applications to Exploratory Multi-way Data Analysis and Blind Source Separations (Wiely, New York, 2009).
M. W. Berry, N. Brown, A. N. Langville, V. P. Pauca, and R. J. Plemmons: Comput. Stat. Data Anal. 52 (2007) 155.
D. D. Lee and H. S. Seung: Adv. Neural Inf. Process. Syst. 13 (2001) 556.
Spectral Database, University of Eastern Finland Color Group [http://spectral.joensuu.fi/].
O. Kohonen, J. Parkkinen, and T. Jääskeläinen: Color Res. Appl. 31 (2006) 381.
R. Ramanath, R. G. Kuehni, W. E. Synder, and D. Hinks: Color Res. Appl. 29 (2004) 29.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Farajikhah, S., Amirshahi, S.H. Initialization of nonnegative matrix factorization by Gaussian primaries for reconstruction of spectral data. OPT REV 19, 294–305 (2012). https://doi.org/10.1007/s10043-012-0046-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10043-012-0046-2