Image Processing Reference
In-Depth Information
ν
≤−
1
. This is done to obtain a better contrast-enhanced image.
The modified non-membership function is given as
IFS
()
g
μ
IFS
()
g
A
A
(
1
+
λμ
)
)
()
(
g
A
1
−
IFS
−
++
1
μ
λμ
()
g
+⋅
+
++
+⋅
1
λμ
g
IFS
IFS
A
ν
() (
g
=
φ μ
( ))
g
=
=
A
A
(
λ
11
1
)(
λ μ
λμ
) ()
()
g
IFS
1
(
1
) ()
g
A
1
(5.14)
g
A
1
−
μ
()
g
A
=
2
13
+⋅⋅
λμ
()
g
+
λ
μ
() ()
g
+
μ
g
A
A
A
IFS
IFS
The hesitation degree is calculated as
π
1 ( ) ( )
.
λ is calculated using the
IF
entropy. The
IF
entropy is calculated as
=− −
μ
g
ν
g
mn
A
A
N
N
1
∑
∑
()
()
1
−
π
g
IE A
()
=
π
ge
Aij
⋅
Aij
N
j
=
1
i
=
1
The optimum value of λ is calculated as
λ
=
max( (;))
IE A
λ
opt
λ
Among all the entropic values for different values of λ, the λ value that corre-
sponds to the maximum entropy is selected. With this λ value, the member-
ship and the hesitation degrees are calculated.
Then fuzzy hedge is applied on the
IF
image which is given as
=
(
()
.
)
125
IFS
IFS
μ
()
g
μ
g
new
A
Contrast stretching is applied on the intuitionistic image using the INT
operator.
The intensifier operation is written as
2
2
⋅
⎣
μ
new
()
g
⎦
if
0
<
μ
new
() .
g
≤
05
IFS
IFS
enh
μ
()
g
=
2
⎣
new
⎦
new
121
−−
μ
()
g
if
05
.
≤
μ
IFS
( )
g
≤
1
IFS
where μ
enh
is the enhanced
IF
image.
5.4.4 Contrast Enhancement by Chaira (Method IV)
This method is suggested by Chaira [7] using Chaira's
IF
generator. The
image is fuzzified with the membership function μ
A
(
g
) using Equation 5.1.
Search WWH ::
Custom Search