Image Processing Reference
In-Depth Information
The optimum value of λ is calculated as
λ
= max( (;))
IE A
λ
opt
λ
The λ value that corresponds to the maximum entropy is selected. With this
λ value, the membership and the hesitation degrees are calculated.
With the λ value, the membership function in Equation 5.15 and the non-
membership function in Equation 5.17 are computed and an IF image is
obtained.
Then a contrast intensifier is applied to the IF image which is written as
2
2
μ
IFS
()
g
if
μ
IFS
() .
g
0 5
A
A
enh
μ
()
g
=
2
IFS
μ IFS ()
121
−−
μ
()
g
if
0
.
5
<
1
A
This enhanced image is the contrast-enhanced image.
Example 5.3
Two examples of medical images using IF methods by Chaira (methods
III and IV) are shown in Figures 5.5 and 5.6 to illustrate the efficacy of
the methods.
5.4.5 Hesitancy Histogram Equalization
In this method, a hesitant histogram is generated, and then using histogram
equalization, image enhancement is obtained.
(a)
(b)
(c)
FIGURE 5.5
(a) CT scan brain (CT-1) image, (b) IF enhancement by Chaira (method III) and (c) the IF
method by Chaira (method IV). (Modified from Chaira, T., Construction of intuitionistic fuzzy
contrast enhanced medical images, in Proc. of IEEE International Conference on Human Computer
Interaction , IIT Kharagpur, India, 2012; Chaira, T., J. Intell. Fuzzy Syst. , 2013.)
Search WWH ::




Custom Search