Image Processing Reference
In-Depth Information
The main part is fuzzification where the grey levels are transformed to the
membership function. The membership function is the degree of belonging-
ness of the pixels in an image. After fuzzification, depending on the user's
requirements, the membership values of the grey level are modified using an
appropriate fuzzy technique.
2.6 Advanced Fuzzy Processing of Medical Images
Till now we know that uncertainties in an image can be dealt better with
fuzzy set theory. But with the introduction of advanced fuzzy set theories
where more or different types of uncertainties are considered, researches
are carried out for obtaining accurate results on medical images. In this sec-
tion, advanced fuzzy techniques such as intuitionistic fuzzy sets and Type II
fuzzy sets are explored for medical image processing.
2.6.1 Intuitionistic Fuzzy Set
As medical images are not equally and well illuminated, for a more accu-
rate diagnosis, advanced fuzzy set theoretic techniques that include intu-
itionistic fuzzy set and Type II fuzzy set are being explored in recent days.
Fuzzy set theory performs better as it considers the vagueness in the form of
a membership function. The membership function is not properly defined
for a particular image. It may be Gaussian, gamma, triangular or any other
function. This is the reason different researchers get varied results with
different membership functions. Membership function selection depends
on the user's choice. So, some kind of hesitation or uncertainty is pres-
ent while defining the membership function. This hesitation is taken into
account in the intuitionistic fuzzy set theory introduced by Atanassov [1].
In fuzzy set, the membership degrees lie between 0 and 1 and the non-
membership degree is the complement of the membership degree. But in
intuitionistic fuzzy set, due to the hesitation degree, the non-membership
degree is not the complement of the membership degree; rather, it is less
than or equal to the complement of the membership degree. It offers an
improved performance as compared to fuzzy set theory. So, in intuition-
istic fuzzy set, two uncertainties are considered: the membership degree
and non-membership degree. With the consideration of more uncertain-
ties, attempts have been taken by several researchers to work on medical
images. As medical images contain uncertainties, intuitionistic fuzzy set
may be useful in medical image processing. The reason is simple: In fuzzy
set, only the membership degrees are considered; non-membership values
Search WWH ::
Custom Search