Premium
Recovery of reflectance spectra from CIE tristimulus values using a progressive database selection technique
Author(s) -
Ansari K,
Amirshahi S H,
Moradian S
Publication year - 2006
Publication title -
coloration technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.297
H-Index - 49
eISSN - 1478-4408
pISSN - 1472-3581
DOI - 10.1111/j.1478-4408.2006.00019.x
Subject(s) - reflectivity , principal component analysis , basis (linear algebra) , series (stratigraphy) , color difference , set (abstract data type) , mathematics , sample (material) , statistics , selection (genetic algorithm) , computer science , pattern recognition (psychology) , artificial intelligence , optics , chemistry , geometry , chromatography , geology , paleontology , enhanced data rates for gsm evolution , physics , programming language
A procedure for creating efficient reflectance spectra from CIE tristimulus colour values is described using a modified linear model. By fixing certain criteria based on colour difference values, the proposed technique preliminarily selects a series of suitable samples from a main data set containing the reflectance values of a large number of different coloured samples, based on the colour specifications of a given sample. In this way, a series of different databases containing the reflectance values of confirmed samples relating to the particular samples are formed. Then, a well‐known principal components linear model is applied using three basis functions. This operation yields higher cumulative variances in comparison with the original database, having the same number of basis vectors. The performance of the proposed method is evaluated using a different collection of samples and the resulting spectra show considerable improvements in terms of root mean square error as well as colour difference values under different illuminants.