Scribd
Upload
947 views10 pages

Color Receipe

This document provides an introduction and overview of color recipe prediction using neuro-fuzzy methodology and computational intelligence. It discusses how color recipe prediction combines fuzzy systems, neural networks, and genetic algorithms as soft computing components. The system takes in 16 input values related to a target color and surface reflectance and outputs 10 colorant concentrations. Key concerns in color recipe prediction are specifying precise colorant levels, limiting the number of colorants used while avoiding complements, accounting for human visual sensitivity, and different colorant combinations producing the same perceptual color. CANFIS (CoActive Neuro-Fuzzy Inference Systems) modeling is introduced for applying neuro-fuzzy models to color recipe prediction. An example of yellow color rules is provided to illustrate the CAN

Uploaded by

Vishesh Gunner Raimugia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
947 views10 pages

Color Receipe

This document provides an introduction and overview of color recipe prediction using neuro-fuzzy methodology and computational intelligence. It discusses how color recipe prediction combines fuzzy systems, neural networks, and genetic algorithms as soft computing components. The system takes in 16 input values related to a target color and surface reflectance and outputs 10 colorant concentrations. Key concerns in color recipe prediction are specifying precise colorant levels, limiting the number of colorants used while avoiding complements, accounting for human visual sensitivity, and different colorant combinations producing the same perceptual color. CANFIS (CoActive Neuro-Fuzzy Inference Systems) modeling is introduced for applying neuro-fuzzy models to color recipe prediction. An example of yellow color rules is provided to illustrate the CAN

Uploaded by

Vishesh Gunner Raimugia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 10

PRESETATION BY:

TEJAS KAVITAKE
DANIEL AKU YIRANG

INTRODUCTION
Color recipe prediction is a practical
application of soft computing.
Color recipe prediction introduces Neurofuzzy methodology and Computational
intelligence
It combines 3 principles of soft computing
components:
(i) Fuzzy systems
(ii) Neural networks
(iii) Genetic algorithms

i/p & o/p relation in color recipe


prediction
16 inputs

10 outputs

Target color

Color
recipe
prediction
system

Surface spectral
reflectance

White
Black
Green 1
Red 1
Green 2
Violet
Red 2
Yellow 1
Yellow 2
Blue

Main concerns in color recipe


prediction

It is difficult to predict precise colorant


concentrations{ we need to specify levels such as
0.01%}.
It is necessary to specify use of limited number of
colorants and we need to avoid the use of
complementary colorants.
The magnitude of mean-squared error of colorant
vectors may not correspond exactly to that of color
differences.
It is important to consider human visual sensitivity
to color difference, which may be costly.
Some different combinations of colorant may have
the same perceptual attributes of color as seen by
humans.

Canfis modeling

CANFIS basically stands for


CoActive Neuro-fuzzy Inference
Systems .

In this section we show how neuro


fuzzy models can be generalized for
application to color recipe prediction.

Understanding basic terms


MFs : Membership Functions
Hue : A perceptual attribute of color
which is linguistic variable
W : Firing strength
{eg : for firing strength of yellow
color we write W(y)}

CANFIS Architectures

Understanding with an example :


YELLOW RULE :
if the target color is yellow, then use yellow rule,
C(y).
YELLOW RULE 1 :
if the target color is greenish yellow then use a
greenish yellow rule, C(gy).
YELLOW RULE 2 :
if the target color is very yellow then use a
very yellow rule, C(vy).
YELLOW RULE 3 :
if the target color is reddish yellow then use a
reddish yellow rule, C(ry).

Canfis with five color rules for color


recipe prediction

Thank you.

You might also like