Indian
Journal of Fibre & Textile Research

Vol.
31, September 2006, pp. 439-443

## Use of transformed reflectance functions for neural network color

match prediction systems

F Ameri^{ }, S Moradian^{ }, M Amani Tehran
& K Faez^{}

Received 19 April 2005; revised
received 26 July 2005; accepted 29 August 2005

Attempts have been
made to use different transformed reflectance functions as input for a fixed
genetically optimized neural network match prediction system. Two different
sets of data depicting dyed samples of known recipes but metameric to each
other were used to train and test the network. All the transformed and
untransformed reflectance functions gave good recipe predictions when trained
and tested by the same data sets (PF/4 being less than 4). However, the
transformation based on matrix R of the decomposition theory showed promising
results, since it gave very good colorant concentration predictions when
trained by the first set of data dyed with one set of colorants while being
tested by a completely different second set of data dyed with a different set
of colorants (PF/ 4 always being less than 10).

**Keywords**:** **Color** **match prediction, Matrix R, Neural
networks, Transformed reflectance functions, Wool

**IPC Code**: Int. Cl.^{8}** **G06N3/02