CONCLUSION
After comparing different mathematical models imposed
on the scattergram, the logistical model was found to
provide a slightly better fit to the experimental data than
the linear in correlating instrumental colour acceptance
determination and visual colour assessment. Even though
the performance of logistical model seems to be preferable
in absolute terms, it is still recommended that a linear
regression line is used when comparing the performance
of different colour-difference equations because it gives a
better overall performance over several batches.
Comparison of the coefficient of determination (R2) and
F ratio using three different colour-difference equations
showed that both the CIE94 and CMC gave a better performance
than CIELAB in terms of individual colour sets.
Although the performance of CIE94 and CMC were quite
similar, it is recommended that the CMC equation is used to
generate microspaces for shade sorting processes since it
gives a better overall performance over a whole set of
batches.
Using the linear mathematical model, the dependent
variable was set as the percentage rejection as assessed by
a panel of 32 observers, the independent variable was
chosen as the colour difference as determined using the
CMC equation. Under these conditions, the 50% rejection
level obtained from the 20 sets of colours used was found
to vary from DE 0.624 to 1.198. These tolerances can be
adopted as the maximum allowable limits for CCC shade
sorting in a microspace generated by the CMC colourdifference
equation.