Design and Implementation - Segmentation of Hand Bone for Bone Age Assessment - page 86

Low medium High

1

0.8

0.6

0.4

0.2

0

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

CER

Fig. 3.16 Membership function of CER

6) Rule 6

Rule antecedent If NDE is far

and CER is low

Rule consequent then Quadruple division Less likely

7) Rule 7

Rule antecedent If NDE is far

and CER is medium

Rule consequent then Quadruple division most likely

For every rule, after evaluation, will produce rule output. The process of com-
bining of all membership function contributed by each rules output into a sin-
gle fuzzy set is called the 'aggregation' process. This process conclude all the
rules output. In Mandani method, this process is accomplished by using the max
operator.
The way human observe, understand, analyze and model is vague and uncer-
tain. However, when it comes to decision making, crisp logic is more reasonable
to human. The decisions human make are mostly in binary nature. For example,
in the context of quadruple division, the final decision is binary: Terminate or
continue the division process. In other words, when human considered the input