I
AE
S
I
nte
rna
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
,
p
p
.
8
~1
7
I
SS
N:
2252
-
8938
,
DOI
: 1
0
.
1
1
5
9
1
/i
j
ai.
v
6
.
i1
.
p
p
8
-
1
7
8
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
jo
u
r
n
a
l.c
o
m/o
n
lin
e/in
d
ex
.
p
h
p
/I
J
AI
CBI
R of Bra
in
MR I
m
a
g
es
Using
H
isto
g
ra
m
of
Fu
zz
y
O
rien
ted
G
ra
dien
ts and
Fu
zz
y
Lo
ca
l Bina
ry
Patter
ns
At
hira
T
R
,
Abra
ha
m
Va
rg
h
ese
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
A
d
i
S
h
a
n
k
a
ra
In
stit
u
te
o
f
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
Ka
lad
y
,
In
d
ia
Art
icle
I
nfo
AB
ST
RA
CT
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
No
v
3
,
2
0
1
6
R
ev
i
s
ed
J
an
7
,
2
0
1
7
A
cc
ep
ted
Feb
16
,
2
0
1
7
Re
tri
e
v
a
l
o
f
si
m
il
a
r
i
m
a
g
e
s
f
ro
m
larg
e
d
a
tas
e
t
o
f
b
ra
in
im
a
g
e
s
a
c
ro
ss
p
a
ti
e
n
ts
w
o
u
ld
h
e
lp
t
h
e
e
x
p
e
rts
in
th
e
d
e
c
isio
n
d
iag
n
o
sis
p
ro
c
e
ss
o
f
d
ise
a
se
s.
G
e
n
e
r
a
ll
y
u
se
d
f
e
a
tu
r
e
e
x
tra
c
ti
o
n
m
e
th
o
d
s
a
re
c
o
lo
r,
tex
tu
re
a
n
d
sh
a
p
e
.
In
m
e
d
ica
l
i
m
a
g
e
s
te
x
tu
re
a
n
d
sh
a
p
e
fe
a
tu
re
s
a
r
e
m
o
st
e
ff
icie
n
t.
His
to
g
ra
m
o
f
Orie
n
ted
G
ra
d
ien
ts
(HO
G
)
a
n
d
L
o
c
a
l
Bin
a
r
y
P
a
tt
e
rn
(L
BP
)
a
re
g
o
o
d
d
e
sc
rip
to
r
f
o
r
b
ra
i
n
M
R
im
a
g
e
re
tri
e
v
a
l.
Bu
t
th
e
re
a
re
m
a
n
y
c
h
a
ll
e
n
g
e
s
f
a
c
in
g
in
m
e
d
ica
l
a
p
p
li
c
a
ti
o
n
.
A
n
e
m
p
iri
c
a
l
stu
d
y
o
f
th
e
i
m
p
a
c
t
o
f
in
c
re
a
sin
g
b
in
s
n
u
m
b
e
r
in
th
e
HO
G
d
e
sc
rip
to
r
c
o
n
c
lu
d
e
d
th
a
t
larg
e
r
th
e
n
u
m
b
e
r
is
m
o
re
a
c
c
u
ra
te
th
e
d
e
s
c
rip
to
r
is.
In
f
a
c
t
th
is
is
d
u
e
to
th
e
re
d
u
c
ti
o
n
o
f
o
rien
tatio
n
s
ra
n
g
e
th
a
t
e
a
c
h
b
in
c
o
v
e
rs.
De
sp
it
e
th
e
e
f
f
icie
n
c
y
o
f
a
u
g
m
e
n
ti
n
g
th
e
b
in
s
n
u
m
b
e
r,
th
is
tec
h
n
iq
u
e
h
a
s
li
m
it
e
d
sp
a
ti
a
l
su
p
p
o
rt
a
s
th
e
a
u
g
m
e
n
tatio
n
o
f
th
e
n
u
m
b
e
r
o
f
b
in
s
u
se
d
lea
d
s
to
in
c
re
a
se
th
e
h
isto
g
ra
m
d
i
m
e
n
sio
n
.
S
o
h
e
re
p
ro
p
o
se
d
a
m
e
th
o
d
c
a
ll
e
d
Hist
o
g
ra
m
o
f
F
u
z
z
y
Or
ien
ted
G
ra
d
ien
ts
(HFOG
),
in
w
h
ich
a
p
ix
e
l
c
a
n
b
e
lo
n
g
se
v
e
ra
l
b
in
s
w
it
h
d
if
f
e
re
n
t
d
e
g
re
e
s.
T
h
e
L
o
c
a
l
Bin
a
ry
P
a
tt
e
rn
s
fe
a
tu
re
e
x
trac
ti
o
n
m
e
th
o
d
is
w
id
e
l
y
u
se
d
f
o
r
te
x
tu
re
a
n
a
l
y
sis;
h
o
w
e
v
e
r,
th
e
o
rig
in
a
l
L
BP
is
b
a
se
d
o
n
h
a
r
d
t
h
re
sh
o
l
d
in
g
t
h
e
n
e
ig
h
b
o
rh
o
o
d
o
f
e
a
c
h
p
ix
e
l.
T
h
e
re
f
o
re
,
tex
tu
re
re
p
re
se
n
tatio
n
w
it
h
L
BP
is
v
e
r
y
s
e
n
siti
v
e
to
n
o
ise
a
n
d
c
a
n
n
o
t
d
isti
n
g
u
ish
b
e
t
w
e
e
n
a
stro
n
g
a
n
d
a
w
e
a
k
p
a
tt
e
rn
.
I
n
t
h
is
stu
d
y
,
F
u
z
z
y
L
o
c
a
l
Bin
a
r
y
P
a
tt
e
rn
s
w
a
s
in
tro
d
u
c
e
d
to
im
p
ro
v
e
th
e
o
rig
in
a
l
L
BP
.
K
ey
w
o
r
d
:
C
B
I
R
Fu
zz
y
lo
ca
l b
in
ar
y
p
atter
n
His
to
g
r
a
m
o
f
f
u
zz
y
o
r
ien
t
ed
g
r
ad
ien
ts
His
to
g
r
a
m
o
f
o
r
ien
ted
g
r
ad
ie
n
ts
L
o
ca
l b
in
ar
y
p
atter
n
Co
p
y
rig
h
t
©
2
0
1
7
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
A
t
h
ir
a
T
R
,
Dep
ar
t
m
en
t o
f
C
o
m
p
u
ter
Sci
e
n
ce
an
d
E
n
g
i
n
ee
r
in
g
,
A
d
i S
h
a
n
k
ar
a
I
n
s
tit
u
te
o
f
E
n
g
i
n
ee
r
in
g
an
d
T
ec
h
n
o
lo
g
y
,
Kala
d
y
,
I
n
d
ia.
E
m
ail:
at
h
ir
atr
0
1
@
g
m
ai
l.c
o
m
1.
I
NT
RO
D
UCT
I
O
N
C
o
n
te
n
t
b
ased
i
m
a
g
e
r
etr
iev
al
p
lay
s
an
i
m
p
o
r
ta
n
t
r
o
le
in
m
a
n
y
ap
p
licatio
n
s
es
p
ec
iall
y
i
n
t
h
e
m
ed
ical
f
ield
.
B
ec
au
s
e
o
f
th
o
u
s
a
n
d
s
o
f
i
m
a
g
es
p
r
o
d
u
ce
d
b
y
t
h
e
h
o
s
p
itals
e
v
er
y
d
a
y
lab
eli
n
g
t
h
e
i
m
ag
e
a
n
d
clas
s
i
f
y
i
n
g
ab
n
o
r
m
alit
y
b
y
h
u
m
a
n
in
ter
v
en
tio
n
i
s
a
ti
m
e
co
n
s
u
m
i
n
g
an
d
d
if
f
icu
lt
tas
k
.
F
u
r
th
er
tr
e
at
m
e
n
t
p
lan
n
i
n
g
i
s
d
ep
en
d
o
n
h
o
w
ac
cu
r
a
tel
y
t
h
e
ab
n
o
r
m
ali
t
y
is
d
etec
ted
,
f
alse
d
etec
tio
n
m
a
y
also
ca
u
s
e
w
r
o
n
g
d
iag
n
o
s
i
s
lead
s
to
s
er
io
u
s
p
r
o
b
lem
to
f
ac
e
t
h
is
s
it
u
atio
n
th
e
b
est
co
n
ten
t
b
ased
i
m
ag
e
r
etr
ie
v
al
m
eth
o
d
is
u
s
ed
to
r
etr
iev
e
q
u
er
y
b
ased
i
m
a
g
e
f
r
o
m
lar
g
e
d
atab
ase
[
1
]
.
A
cc
u
r
ate
r
es
u
lt
ca
n
b
e
p
r
o
d
u
ce
d
o
n
l
y
w
it
h
th
e
h
elp
o
f
a
u
to
m
ated
co
m
p
u
ter
aid
ed
tech
n
iq
u
e
[
2
]
.
C
o
n
ten
t
B
ased
Me
d
ical
I
m
ag
e
R
etr
iev
al
(
C
B
MI
R
)
[
3
]
ca
n
b
e
u
s
ef
u
l
f
o
r
m
a
n
y
d
is
ea
s
es
s
u
c
h
as
b
r
ain
t
u
m
o
r
,
b
r
ea
s
t
ca
n
ce
r
,
s
p
in
e
d
is
o
r
d
er
p
r
o
b
lem
etc
w
h
ic
h
i
s
ac
q
u
i
r
ed
th
r
o
u
g
h
m
a
n
y
m
o
d
a
li
ties
s
u
c
h
as C
T
s
ca
n
,
M
R
I
,
m
a
m
m
o
g
r
a
m
etc.
T
h
e
m
aj
o
r
co
n
ten
t
in
i
m
ag
e
s
u
s
ed
co
n
s
is
ts
o
f
co
lo
r
[
4
]
,
s
h
ap
e
[
5
]
an
d
tex
tu
r
e
f
ea
tu
r
es
[
6
]
.
C
o
lo
r
h
is
to
g
r
a
m
is
t
h
e
co
m
m
o
n
l
y
u
s
ed
m
e
th
o
d
f
o
r
co
lo
r
f
ea
tu
r
e
ex
tr
ac
tio
n
i
n
d
ig
ital
i
m
ag
e
s
.
B
u
t
it
is
ex
tr
e
m
e
l
y
d
if
f
ic
u
l
t
to
h
a
n
d
le
f
o
r
co
m
p
u
t
er
s
,
b
ec
au
s
e
th
e
y
ex
tr
ac
t
t
h
e
c
o
lo
r
in
f
o
r
m
atio
n
f
r
o
m
a
n
i
m
a
g
e
w
i
th
o
u
t
co
n
tex
t
in
f
o
r
m
atio
n
.
So
s
h
ap
e
an
d
tex
tu
r
e
f
ea
tu
r
e
s
ar
e
m
o
r
e
s
u
ita
b
le
f
o
r
im
a
g
e
r
etr
iev
al
ap
p
licatio
n
s
esp
ec
iall
y
i
n
m
ed
ical
i
m
ag
e
r
etr
ie
v
al.
T
h
er
e
ar
e
m
a
in
l
y
t
w
o
t
y
p
e
s
o
f
s
h
a
p
e
m
a
tch
i
n
g
tech
n
iq
u
es.
T
h
e
y
ar
e
g
eo
m
etr
y
b
ased
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
B
I
R
o
f B
r
a
in
MR
I
ma
g
es Usin
g
His
to
g
r
a
m
o
f F
u
z
z
y
Ori
en
ted
Gra
d
ien
ts
a
n
d
…
(
A
th
ir
a
TR
)
9
an
d
s
tr
u
ct
u
r
al
b
ased
s
h
ap
e
m
atch
i
n
g
.
T
h
er
e
ar
e
m
a
n
y
m
et
h
o
d
s
f
o
r
tex
tu
r
e
f
ea
t
u
r
e
ex
tr
a
ctio
n
.
B
ased
o
n
th
e
d
o
m
ai
n
f
r
o
m
w
h
ich
t
h
e
tex
tu
r
e
f
ea
tu
r
e
is
e
x
tr
ac
ted
,
th
ese
m
eth
o
d
s
ar
e
b
r
o
ad
ly
class
i
f
ied
i
n
t
o
t
w
o
ca
te
g
ir
ies.
T
h
ey
ar
e
s
p
ac
ial
tex
tu
r
e
f
ea
tu
r
e
ex
tr
ac
tio
n
m
et
h
o
d
an
d
Sp
ec
tr
al
tex
tu
r
e
f
ea
tu
r
e
e
x
tr
ac
tio
n
m
eth
o
d
s
.
I
n
s
p
atia
l
ap
p
r
o
ac
h
,
tex
tu
r
e
f
ea
t
u
r
es
ar
e
ex
tr
ac
ted
b
y
co
m
p
u
t
in
g
t
h
e
p
ix
el
s
tat
is
tic
s
o
r
f
i
n
d
in
g
th
e
lo
ca
l
p
ix
el
s
tr
u
ctu
r
e
s
in
o
r
ig
i
n
al
i
m
a
g
e
d
o
m
ain
.
T
h
e
s
p
atial
te
x
t
u
r
e
f
ea
t
u
r
e
ex
t
r
ac
tio
n
tech
n
iq
u
e
s
ca
n
b
e
f
u
r
th
er
class
i
f
ied
a
s
s
tr
u
ct
u
r
al,
s
tati
s
tical
an
d
m
o
d
el
b
ased
.
I
n
s
p
ec
tr
al
tex
tu
r
e
f
ea
tu
r
e
ex
tr
ac
tio
n
tech
n
iq
u
es,
an
i
m
a
g
e
is
tr
an
s
f
o
r
m
ed
in
to
f
r
eq
u
en
c
y
d
o
m
ai
n
an
d
th
e
n
f
ea
t
u
r
e
is
ca
lcu
l
a
ted
f
r
o
m
th
e
tr
a
n
s
f
o
r
m
ed
i
m
a
g
e.
T
h
er
e
ar
e
m
an
y
d
if
f
ic
u
ltie
s
f
a
cin
g
w
h
i
le
d
ea
lin
g
w
it
h
r
ea
l
b
r
ain
MR
i
m
a
g
es,
d
u
e
to
ch
a
ll
en
g
e
s
li
k
e
MR
i
n
ter
-
a
n
d
in
tr
a
-
p
atie
n
t
i
n
ten
s
i
t
y
v
ar
iatio
n
s
,
m
is
al
ig
n
m
en
t
o
f
i
m
a
g
es,
etc.
An
e
m
p
ir
ical
s
t
u
d
y
o
f
HOG
co
n
clu
d
es
t
h
at
w
h
e
n
t
h
e
n
u
m
b
er
o
f
b
in
s
in
cr
ea
s
es
HOG
[
7
]
[
8
]
w
ill
b
ec
o
m
e
m
o
r
e
ac
cu
r
ate.
B
u
t
it
is
li
m
ited
s
p
atial
s
u
p
p
o
r
t
as
th
e
au
g
m
e
n
t
atio
n
o
f
t
h
e
n
u
m
b
er
o
f
b
i
n
s
u
s
ed
lead
s
to
i
n
cr
ea
s
e
th
e
h
is
to
g
r
a
m
d
i
m
en
s
io
n
.
So
h
er
e
p
r
o
p
o
s
ed
a
m
eth
o
d
ca
lle
d
His
to
g
r
a
m
o
f
F
u
zz
y
Or
ie
n
te
d
Gr
ad
ien
ts
(
HFOG)
[
9
]
.
T
h
er
e
ca
n
also
in
cr
ea
s
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
L
B
P
[
1
0
]
b
y
t
h
e
co
n
ce
p
t
o
f
f
u
zz
y
a
p
p
lied
to
it,
ca
lled
Fu
zz
y
L
o
ca
l
B
in
ar
y
P
atter
n
(
FL
B
P
)
.
Fu
zz
if
icatio
n
allo
w
s
a
Fu
zz
y
L
o
ca
l
B
in
ar
y
P
atter
n
(
FL
B
P
)
[
1
1
]
to
co
n
tr
ib
u
te
to
m
o
r
e
t
h
an
a
s
in
g
l
e
b
in
in
th
e
d
i
s
tr
ib
u
tio
n
o
f
t
h
e
L
B
P
v
alu
es
u
s
ed
as
a
f
ea
t
u
r
e
v
e
cto
r
.
A
f
u
s
io
n
m
et
h
o
d
o
f
HFO
G
an
d
F
L
B
P
u
s
ed
f
o
r
in
cr
ea
s
in
g
th
e
ac
cu
r
ac
y
o
f
r
etr
iev
al.
T
h
e
f
u
s
ed
f
ea
t
u
r
es
ar
e
p
u
tted
i
n
to
SVM
clas
s
i
f
ier
f
o
r
lear
n
in
g
.
H
OG
f
ea
t
u
r
es
o
p
er
ate
in
t
h
e
lo
ca
l
ce
ll
u
n
it
o
f
th
e
i
m
a
g
e,
s
o
i
t
less
s
en
s
iti
v
e
to
t
h
e
g
eo
m
et
r
ic
tr
an
s
f
o
r
m
atio
n
,
r
o
tatio
n
an
d
th
e
illu
m
i
n
atio
n
ch
an
g
es
in
t
h
e
i
m
a
g
e.
B
u
t
HOG
p
er
f
o
r
m
s
p
o
o
r
ly
w
h
e
n
th
e
b
ac
k
g
r
o
u
n
d
is
clu
tter
ed
w
ith
n
o
is
y
ed
g
es.
L
o
ca
l
B
in
ar
y
P
atter
n
i
s
co
m
p
le
m
en
tar
y
i
n
th
i
s
asp
ec
t.
L
B
P
f
ea
tu
r
es
h
a
v
e
q
u
ic
k
ca
lcu
latio
n
s
p
ee
d
,
g
o
o
d
r
o
b
u
s
tn
es
s
to
lig
h
t.
B
u
t
b
ec
au
s
e
t
h
e
s
ize
o
f
L
B
P
w
i
n
d
o
w
is
f
i
x
ed
an
d
h
as
n
o
th
i
n
g
to
d
o
w
ith
th
e
i
m
a
g
e,
L
B
P
m
a
y
h
av
e
er
r
o
r
s
in
t
h
e
tex
tu
r
e
c
h
ar
ac
ter
is
tics
e
x
tr
ac
tio
n
,
s
o
it is
d
i
f
f
icu
l
t to
s
ati
s
f
y
t
h
e
r
eq
u
ir
e
m
en
ts
o
f
d
i
f
f
er
e
n
t
r
o
u
g
h
n
es
s
an
d
s
ca
le
te
x
t
u
r
e.
Fu
s
io
n
m
et
h
o
d
w
ill
r
e
s
u
lt
b
etter
p
er
f
o
r
m
a
n
ce
an
d
ac
cu
r
ac
y
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
Her
e
w
e
h
a
v
e
d
o
n
e
t
w
o
m
et
h
o
d
s
His
to
g
r
a
m
o
f
F
u
zz
y
Or
ie
n
ted
Gr
ad
ien
t
s
a
n
d
Fu
zz
y
L
o
ca
l
B
in
ar
y
P
atter
n
s
.
S
h
o
w
n
in
Fig
u
r
e
1
.
Fig
u
r
e
1
.
C
B
I
R
S
y
s
t
em
2
.
1
.
H
is
t
o
g
ra
m
o
f
F
uzzy
O
rient
ed
G
ra
die
nt
(
H
F
O
G
)
His
to
g
r
a
m
o
f
o
r
ien
ted
g
r
ad
ien
t
(
HOG)
f
ea
t
u
r
e
d
escr
ip
to
r
s
w
a
s
p
r
o
p
o
s
ed
b
y
Dala
l
a
n
d
T
r
ig
g
s
[
7
]
.
T
h
e
id
ea
b
eh
in
d
HOG
d
escr
ip
to
r
is
b
ased
o
n
ed
g
e
in
f
o
r
m
atio
n
.
T
h
e
HOG
d
escr
ip
to
r
tech
n
iq
u
e
co
u
n
t
s
o
cc
u
r
r
en
c
es
o
f
g
r
ad
ien
t
o
r
ie
n
tatio
n
i
n
lo
ca
lized
p
o
r
tio
n
s
o
f
an
i
m
ag
e
d
etec
tio
n
w
i
n
d
o
w
,
o
r
r
eg
io
n
o
f
in
ter
est (
R
OI
)
[
8
]
.
T
h
e
HFOG
[
9
]
s
h
o
u
ld
r
ef
lect
a
m
o
r
e
o
b
j
ec
tiv
e
d
escr
ip
tio
n
o
f
co
n
ten
t
t
h
a
n
t
h
e
co
n
v
e
n
ti
o
n
al
HO
G
f
ea
t
u
r
es.
T
h
e
m
ain
s
tep
s
s
ta
y
s
s
a
m
e
a
s
HO
G
i
n
F
-
HO
G.
First,
t
h
e
i
m
a
g
e
d
er
iv
at
iv
e
in
th
e
h
o
r
izo
n
tal
an
d
v
er
tical
d
ir
ec
tio
n
s
ar
e
ca
lcu
la
ted
,
n
ex
t
s
tep
co
n
s
i
s
ts
i
n
co
m
p
u
ti
n
g
th
e
g
r
ad
ien
t
an
d
th
e
m
a
g
n
it
u
d
e
o
f
ea
ch
p
ix
el
i
n
th
e
i
m
ag
e.
T
h
en
,
co
n
t
r
ar
ily
to
th
e
s
ta
n
d
ar
d
HOG
w
h
en
ea
ch
o
n
e
o
f
th
e
s
e
p
ix
el
s
v
o
tes f
o
r
a
u
n
iq
u
e
b
i
n
b
y
ad
d
in
g
it
s
m
a
g
n
i
tu
d
e.
He
r
e
b
elo
n
g
i
n
g
o
f
a
p
i
x
el
i
n
a
b
in
i
s
n
o
t
o
b
s
o
lete.
I
n
t
h
is
co
n
tex
t,
th
er
e
w
il
l
ass
o
ciate
to
ea
ch
p
ix
el
a
f
u
zz
y
d
escr
ip
tio
n
co
n
s
id
er
in
g
m
o
r
e
th
an
o
n
e
b
in
;
t
h
is
r
ep
r
esen
t
s
th
e
co
n
tr
ib
u
tio
n
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
8
–
17
10
th
e
p
i
x
el
i
n
to
ea
c
h
o
n
e
o
f
t
h
es
e
b
in
s
.
L
e
t
I
(
x
,
y
)
b
e
a
p
i
x
el
with
a
m
ag
n
it
u
d
e
M(
x
,
y
)
a
n
d
a
g
r
ad
ien
t
(
x
,
y
)
.
I
n
th
e
o
r
ig
i
n
al
HOG,
I
(
x
,
y
)
w
ill
b
elo
n
g
o
n
l
y
to
th
e
r
an
g
e
[
]
w
h
e
r
e
k
is
a
p
o
s
itiv
e
in
te
g
er
b
elo
n
g
i
n
g
to
[
0
;
n
]
an
d
n
is
t
h
e
n
u
m
b
er
o
f
b
in
s
co
n
s
id
er
ed
.
I
n
an
o
th
er
wo
r
d
s
,
th
e
w
h
o
le
m
a
g
n
it
u
d
e
o
f
th
e
p
ix
el
I
(
x
;
y
)
w
il
l
b
e
ad
d
ed
o
n
l
y
to
t
h
e
r
an
g
e
[
]
.
Ho
w
e
v
er
,
in
th
e
p
r
o
p
o
s
ed
HFOG,
b
ased
o
n
t
h
e
id
ea
t
h
at
a
p
ix
el
g
r
ad
ien
t
ca
n
b
elo
n
g
to
s
e
v
er
al
b
in
s
at
t
h
e
s
a
m
e
ti
m
e,
t
h
e
m
ag
n
i
tu
d
e
o
f
I
(
x
;
y
)
w
i
ll
b
e
ad
d
ed
to
all
t
h
ese
b
i
n
s
w
it
h
a
d
eg
r
ee
s
p
ec
i
f
ied
b
y
m
e
m
b
er
s
h
ip
g
r
ad
es
b
et
w
ee
n
0
an
d
1
.
Fo
r
th
is
p
u
r
p
o
s
e,
a
m
e
m
b
er
s
h
ip
d
eg
r
ee
i
s
d
ef
in
ed
u
s
i
n
g
t
h
e
F
u
zz
y
m
e
m
b
er
s
h
ip
f
u
n
ct
io
n
o
f
F
u
zz
y
-
C
-
Me
an
s
al
g
o
r
ith
m
:
(
)
∑
(
‖
‖
‖
‖
)
W
h
er
e
(
)
is
th
e
co
n
tr
ib
u
tio
n
o
f
th
e
p
ix
el
i
n
in
t
h
e
v
al
u
e
o
f
t
h
e
b
in
,
an
d
m
is
a
p
ar
a
m
eter
w
h
ic
h
co
n
tr
o
ls
th
e
d
e
g
r
ee
o
f
f
u
zz
in
e
s
s
in
th
e
d
i
s
tr
ib
u
tio
n
o
f
p
ix
els
(
1
<
m
<1
)
.
E
m
p
ir
icall
y
,
f
o
u
n
d
th
a
t
m
=1
.
1
is
th
e
b
est
s
et
tin
g
.
O
n
ce
t
h
e
d
eg
r
ee
o
f
v
o
tin
g
o
f
a
p
ix
el
o
n
ea
ch
b
in
i
s
ca
lc
u
lated
,
th
e
m
a
g
n
i
tu
d
e
o
f
th
is
p
ix
e
l
w
il
l b
e
ad
d
ed
to
th
e
b
in
s
,
d
ep
en
d
in
g
to
th
e
m
e
m
b
er
s
h
ip
d
eg
r
ee
.
A
.
A
l
g
o
r
ith
m
:
F
u
zz
y
L
o
ca
l B
i
n
ar
y
P
atter
n
I
n
p
u
t: Q
u
er
y
I
m
ag
e
Ou
tp
u
t: HFOG
f
ea
t
u
r
e
v
ec
to
r
Step
s
:
1.
C
alcu
late
th
e
i
m
a
g
e
g
r
ad
ie
n
t i
n
X
an
d
Y
d
ir
ec
tio
n
s
.
2.
C
o
m
p
u
te
th
e
g
r
ad
ien
t
an
d
m
a
g
n
i
tu
d
e
o
f
ea
ch
p
ix
el
i
n
th
e
i
m
ag
e.
3.
E
ac
h
o
n
e
o
f
t
h
ese
p
ix
e
ls
v
o
te
s
f
o
r
a
u
n
iq
u
e
b
in
ac
co
r
d
in
g
to
th
e
o
r
ien
tatio
n
o
f
t
h
e
p
ix
el.
i.e
.
,
(
)
v
alu
e
s
.
4.
C
alcu
late
th
e
a
v
er
ag
e
o
f
ea
c
h
b
in
,
i.e
.
,
.
5.
C
alcu
lat
e
co
n
tr
ib
u
t
io
n
o
f
ea
c
h
an
g
le
i
n
t
h
e
r
esp
ec
tiv
e
B
in
s
,
(
)
.
6.
Mu
ltip
l
y
(
)
w
it
h
its
m
ag
n
it
u
d
e
M.
7.
Scale
th
e
i
m
a
g
e
in
t
h
e
r
an
g
e
(
0
,
2
5
0
)
.
8.
Fin
d
t
h
e
n
o
r
m
al
ized
h
is
to
g
r
a
m
w
h
ich
i
s
th
e
f
ea
t
u
r
e
v
ec
to
r
o
f
th
e
i
m
a
g
e.
9.
P
r
o
ce
s
s
is
r
ep
ea
ted
f
o
r
all
im
a
g
es
i
n
th
e
d
atab
ase
a
n
d
f
ea
t
u
r
e
d
atab
ase
is
f
o
r
m
.
10.
P
er
f
o
r
m
s
i
m
i
lar
it
y
co
m
p
u
ti
n
g
.
2
.
2
.
F
uzzy
L
o
ca
l B
ina
ry
P
a
t
t
er
ns
(
L
B
P
)
L
o
ca
l
b
in
ar
y
p
atter
n
s
w
er
e
in
tr
o
d
u
ce
d
b
y
Oj
ala
et
al
[
1
0
]
as
a
f
i
n
e
s
ca
le
tex
tu
r
e
d
escr
ip
to
r
.
L
o
ca
l
B
in
ar
y
P
atter
n
(
L
B
P
)
[
1
0
]
is
a
s
i
m
p
le
y
et
v
er
y
e
f
f
icie
n
t
te
x
tu
r
e
o
p
er
ato
r
w
h
ic
h
lab
els
t
h
e
p
ix
els
o
f
a
n
i
m
a
g
e
b
y
t
h
r
es
h
o
ld
in
g
t
h
e
n
ei
g
h
b
o
r
h
o
o
d
o
f
ea
ch
p
ix
el
an
d
co
n
s
id
er
s
th
e
r
esu
l
t
as
a
b
in
ar
y
n
u
m
b
er
.
Du
e
to
its
d
is
cr
i
m
i
n
ati
v
e
p
o
w
er
an
d
co
m
p
u
tat
io
n
al
s
i
m
p
licit
y
,
L
B
P
tex
tu
r
e
o
p
er
ato
r
h
as
b
ec
o
m
e
a
p
o
p
u
lar
ap
p
r
o
ac
h
in
v
ar
io
u
s
ap
p
licatio
n
s
.
I
t
ca
n
b
e
s
ee
n
as
a
u
n
i
f
y
i
n
g
ap
p
r
o
ac
h
to
th
e
tr
ad
itio
n
all
y
d
iv
er
g
en
t
s
tati
s
tical
an
d
s
tr
u
ct
u
r
al
m
o
d
els
o
f
tex
tu
r
e
a
n
al
y
s
i
s
.
P
er
h
ap
s
th
e
m
o
s
t
i
m
p
o
r
tan
t
p
r
o
p
e
r
ty
o
f
t
h
e
L
B
P
o
p
er
ato
r
in
r
ea
l
-
w
o
r
ld
ap
p
licatio
n
s
is
its
r
o
b
u
s
t
n
es
s
to
m
o
n
o
to
n
ic
g
r
a
y
-
s
ca
le
ch
an
g
es
ca
u
s
ed
,
f
o
r
ex
a
m
p
l
e,
b
y
ill
u
m
in
at
io
n
v
ar
iatio
n
s
.
An
o
t
h
er
i
m
p
o
r
tan
t
p
r
o
p
er
ty
is
it
s
co
m
p
u
tatio
n
al
s
i
m
p
licit
y
,
w
h
ich
m
a
k
es
it
p
o
s
s
ib
le
to
an
al
y
z
e
i
m
a
g
es i
n
ch
al
len
g
i
n
g
r
ea
l
-
ti
m
e
s
ettin
g
s
.
L
B
P
d
escr
ip
tio
n
o
f
a
p
ix
el
is
cr
ea
ted
b
y
th
r
es
h
o
ld
in
g
.
T
h
e
v
alu
e
s
o
f
th
e
3
X3
n
eig
h
b
o
r
h
o
o
d
o
f
th
e
p
ix
el
ag
ai
n
s
t th
e
ce
n
tr
al
p
ix
el
an
d
in
ter
p
r
etin
g
t
h
e
r
esu
l
t a
s
a
b
in
ar
y
n
u
m
b
er
.
T
h
e
o
r
ig
in
al
L
B
P
is
b
ased
o
n
h
ar
d
th
r
es
h
o
ld
in
g
t
h
e
n
ei
g
h
b
o
r
h
o
o
d
o
f
ea
ch
p
i
x
el,
w
h
i
ch
m
ak
e
s
tex
t
u
r
e
r
ep
r
esen
tatio
n
s
e
n
s
i
tiv
e
to
n
o
is
e.
I
n
ad
d
itio
n
,
L
B
P
ca
n
n
o
t
d
is
ti
n
g
u
is
h
b
et
w
ee
n
a
s
tr
o
n
g
a
n
d
a
w
ea
k
p
atter
n
.
I
n
o
r
d
er
to
en
h
an
ce
t
h
e
L
B
P
ap
p
r
o
ac
h
[
1
1
]
,
Fu
zz
y
L
o
ca
l
B
in
ar
y
P
a
tter
n
s
(
F
L
B
P
)
[
1
2
]
is
p
r
o
p
o
s
ed
.
I
n
F
L
B
P
,
an
y
n
ei
g
h
b
o
r
h
o
o
d
d
o
es n
o
t r
ep
r
esen
ted
o
n
l
y
b
y
o
n
e
co
d
e,
b
u
t,
it is
r
ep
r
esen
ted
b
y
all
e
x
i
s
ti
n
g
co
d
es
w
it
h
d
if
f
er
en
t
d
eg
r
ee
s
.
I
n
F
L
B
P
,
an
y
f
u
zz
y
I
n
ter
s
ec
tio
n
a
n
d
U
n
io
n
o
p
er
ato
r
s
m
a
y
b
e
u
s
ed
.
S
h
o
w
n
in
F
ig
u
r
e
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
B
I
R
o
f B
r
a
in
MR
I
ma
g
es Usin
g
His
to
g
r
a
m
o
f F
u
z
z
y
Ori
en
ted
Gra
d
ien
ts
a
n
d
…
(
A
th
ir
a
TR
)
11
(
a)
(
b
)
Fig
u
r
e
2
.
Qu
er
y
I
m
a
g
e
a
n
d
its
C
o
r
r
esp
o
n
d
in
g
L
B
P
P
atter
n
T
h
e
f
u
zz
i
f
icat
io
n
o
f
th
e
L
B
P
ap
p
r
o
ac
h
[
1
2
]
in
clu
d
es
t
h
e
t
r
an
s
f
o
r
m
atio
n
o
f
t
h
e
i
n
p
u
t
v
a
r
iab
les
to
r
esp
ec
tiv
e
f
u
zz
y
v
ar
iab
les,
ac
co
r
d
in
g
to
a
s
et
o
f
f
u
zz
y
r
u
le
s
.
Fo
r
th
is
p
u
r
p
o
s
e,
t
w
o
f
u
zz
y
r
u
les
ar
e
i
n
tr
o
d
u
ce
d
to
d
escr
ib
e
th
e
r
elatio
n
b
et
wee
n
t
h
e
i
n
te
n
s
i
t
y
v
al
u
es
o
f
t
h
e
n
eig
h
b
o
r
h
o
o
d
p
ix
els
an
d
th
e
ce
n
tr
al
p
ix
el
as f
o
llo
w
s
:
A
cc
o
r
d
in
g
to
th
e
ab
o
v
e
r
u
les,
t
w
o
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
s
,
m
0
an
d
m
1
,
ar
e
n
ee
d
ed
.
L
et
f
u
n
ct
io
n
s
m
0
an
d
m
1
d
ef
in
e
t
h
e
d
eg
r
ee
to
w
h
ic
h
is
0
an
d
1
,
r
esp
ec
tiv
el
y
.
T
h
e
m
0
s
h
o
u
ld
b
e
d
escen
d
in
g
with
r
esp
ec
t
to
.
T
h
er
ef
o
r
e,
as a
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
m
0
,
w
e
co
n
s
id
er
th
e
d
ec
r
ea
s
in
g
f
u
n
ctio
n
d
ef
in
ed
i
n
eq
u
atio
n
b
elo
w
:
(
)
{
On
t
h
e
o
th
er
h
a
n
d
,
m
e
m
b
er
s
h
ip
f
u
n
ctio
n
m
1
d
ef
i
n
e
s
th
e
d
eg
r
ee
to
w
h
ich
is
1
.
T
h
is
f
u
n
c
tio
n
m
a
y
b
e
t
h
e
co
m
p
le
m
e
n
t o
f
$
m
0.
i.e
.
,
(
)
(
)
Fo
r
b
o
th
m
0
a
n
d
m
1
,
T
[
0
,
2
5
5
]
r
ep
r
esen
ts
a
p
ar
a
m
eter
t
h
at
co
n
tr
o
ls
th
e
d
eg
r
ee
o
f
f
u
zz
in
e
s
s
.
Fig
u
r
e
3
.
Me
m
b
e
r
s
h
ip
F
u
n
ctio
n
s
m
0
a
n
d
m
1
as a
F
u
n
ctio
n
o
f
P
i
–
P
ce
n
tr
e
A
lt
h
o
u
g
h
f
o
r
th
e
o
r
ig
i
n
al
L
B
P
o
p
er
ato
r
a
s
in
g
le
L
B
P
co
d
e
ch
ar
ac
ter
izes
a
n
ei
g
h
b
o
r
h
o
o
d
,
b
u
t,
in
t
h
e
p
r
o
p
o
s
ed
FL
B
P
ap
p
r
o
ac
h
,
a
n
eig
h
b
o
r
h
o
o
d
m
a
y
b
e
c
h
ar
ac
ter
ized
b
y
all
av
ailab
le
L
B
P
co
d
es
o
n
d
i_
er
en
t
m
e
m
b
er
s
h
ip
d
eg
r
ee
s
.
T
h
e
m
e
m
b
er
s
h
ip
d
eg
r
ee
o
f
ea
c
h
L
B
P
co
d
e
in
a
p
ix
el
d
ep
en
d
s
o
n
t
h
e
m
e
m
b
er
s
h
i
p
f
u
n
ctio
n
s
m
0
an
d
m1
s
h
o
w
n
i
n
Fi
g
u
r
e
3
,
an
d
all
n
ei
g
h
b
o
r
h
o
o
d
v
alu
e
s
o
f
t
h
e
p
ix
el.
Fo
r
an
y
n
eig
h
b
o
r
h
o
o
d
,
th
e
co
n
tr
ib
u
tio
n
C
L
B
P
o
f
ea
ch
L
B
P
co
d
e
is
d
e
f
i
n
ed
as:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
8
–
17
12
⋂
(
)
w
h
er
e
[
]
is
a
L
B
P
co
d
e
an
d
is
th
e
b
it o
f
it.
A
ls
o
,
is
t
h
e
in
ter
s
ec
tio
n
(
AND)
o
p
er
ato
r
.
A
f
ter
co
m
p
u
ti
n
g
t
h
e
co
n
tr
ib
u
tio
n
o
f
th
e
L
B
P
co
d
es in
all
o
f
t
h
e
p
ix
el
s
,
a
h
is
to
g
r
a
m
ca
n
b
e
d
ef
i
n
ed
as:
⋃
(
)
w
h
er
e
is
th
e
u
n
io
n
o
p
er
ato
r
.
T
h
is
FL
B
P
h
i
s
to
g
r
a
m
co
n
tai
n
s
in
f
o
r
m
at
io
n
ab
o
u
t
t
h
e
d
is
tr
i
b
u
tio
n
o
f
th
e
lo
ca
l
m
icr
o
p
atter
n
s
,
s
u
c
h
a
s
ed
g
e
s
,
s
p
o
ts
an
d
f
lat
ar
ea
s
o
v
er
t
h
e
w
h
o
le
i
m
a
g
e,
s
o
ca
n
b
e
u
s
ed
to
s
tati
s
ticall
y
d
escr
ib
e
th
e
i
m
a
g
e
ch
ar
ac
ter
is
tics
.
B
.
A
lg
o
r
it
h
m
:
F
u
zz
y
L
o
ca
l B
in
ar
y
P
atter
n
I
n
p
u
t: Q
u
er
y
I
m
ag
e
Ou
tp
u
t: F
L
B
P
f
ea
tu
r
e
v
ec
to
r
S
en
ten
ce
_
T
DW
_
Featu
r
e
=P
r
ep
r
o
ce
s
s
in
g
(
W
)
;
Step
s
:
1.
Dete
r
m
i
n
e
th
e
d
i
m
e
n
s
io
n
s
o
f
t
h
e
in
p
u
t i
m
ag
e.
2.
B
lo
ck
s
ize,
ea
ch
L
B
P
co
d
e
is
co
m
p
u
ted
w
it
h
in
a
b
lo
ck
o
f
s
i
ze
b
s
ize
y
*
b
s
izex
3.
C
alcu
late
d
x
an
d
d
y
.
d
x
=
x
s
iz
e
-
b
s
ize
x
,
d
y
=
y
s
ize
-
b
s
ize
y
.
4.
Fil
l th
e
ce
n
tr
al
p
ix
el
m
atr
i
x
C
.
5.
I
n
itialize
t
h
e
r
esu
lt
m
a
tr
ix
w
i
t
h
ze
r
o
s
.
6.
C
o
m
p
u
te
th
e
n
ei
g
h
b
o
u
r
i
m
ag
e
u
s
i
n
g
t
h
e
r
u
le
s
R
0
:
.
.
7.
C
alcu
late
m
0
a
n
d
m
1,
(
)
{
an
d
(
)
(
)
.
8.
Fo
r
all
n
eig
h
b
o
u
r
h
o
o
d
s
ass
ig
n
co
n
tr
ib
u
tio
n
to
h
is
to
g
r
a
m
.
⋂
(
)
9.
C
alcu
late
th
e
h
is
to
g
r
a
m
,
w
h
ic
h
is
t
h
e
f
ea
t
u
r
e
v
ec
to
r
o
f
th
e
i
m
ag
e.
⋃
(
)
10.
P
r
o
ce
s
s
is
r
ep
ea
ted
f
o
r
all
im
a
g
es i
n
th
e
d
atab
ase
a
n
d
f
ea
t
u
r
e
d
atab
ase
is
f
o
r
m
.
11.
P
er
f
o
r
m
s
i
m
i
lar
it
y
co
m
p
u
ti
n
g
.
2
.
3
.
Si
m
ila
rit
y
M
e
a
s
ure
Fo
r
co
m
p
ar
in
g
th
e
s
i
m
ilar
it
y
o
f
t
w
o
m
e
tr
ic
w
e
u
s
e
E
u
clid
e
an
d
is
tan
ce
.
I
n
C
ar
te
s
ian
co
o
r
d
in
ates
,
if
p
=
(
p
1
,
p
2
,
.
.
.
,
p
n
)
an
d
q
=
(
q
1
,
q
2
,...,
q
n
)
ar
e
t
w
o
p
o
in
ts
in
E
u
clid
ea
n
n
-
s
p
ac
e
,
th
e
n
t
h
e
d
is
ta
n
ce
(
d
)
f
r
o
m
p
to
q
,
o
r
f
r
o
m
q
to
p
i
s
g
i
v
e
n
b
y
t
h
e
P
y
t
h
a
g
o
r
ea
n
f
o
r
m
u
la
:
(
)
√
∑
(
)
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
B
I
R
o
f B
r
a
in
MR
I
ma
g
es Usin
g
His
to
g
r
a
m
o
f F
u
z
z
y
Ori
en
ted
Gra
d
ien
ts
a
n
d
…
(
A
th
ir
a
TR
)
13
2
.
4
.
P
er
f
o
rm
a
nce
E
v
a
lua
t
io
n
Av
er
a
g
e
Ra
nk
:
Av
er
a
g
e
r
an
k
is
ca
lc
u
lated
u
s
i
n
g
t
h
e
i
m
a
g
es
r
etr
iev
ed
b
ased
o
n
a
s
et
o
f
r
an
d
o
m
n
u
m
b
er
o
f
q
u
er
y
i
m
a
g
es.
I
n
p
er
f
ec
t
ca
s
e
its
v
al
u
e
i
s
1
i.e
.
,
r
elev
an
t
i
m
ag
e
s
ar
e
i
n
a
s
u
cc
ee
d
in
g
o
r
d
er
w
it
h
o
u
t
th
e
p
r
esen
ce
o
f
ir
r
ele
v
an
t i
m
a
g
es i
n
b
et
w
ee
n
t
h
e
m
.
A
v
er
a
g
e
r
an
k
is
ca
lc
u
lated
u
s
i
n
g
t
h
e
f
o
r
m
u
la
:
(
∑
(
)
)
3.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
I
n
o
u
r
ex
p
er
i
m
e
n
t,
d
if
f
er
en
t
t
y
p
es
o
f
ax
ial
v
ie
w
o
f
b
r
ain
i
m
ag
e
d
ataset
s
ar
e
u
s
ed
,
w
h
ich
i
s
d
o
w
n
lo
ad
ed
f
r
o
m
p
u
b
licl
y
a
v
ailab
le
B
r
ain
W
eb
d
atase
t
[
h
ttp
://b
r
ain
w
eb
.
b
ic.
m
n
i.
m
c
g
ill
.
ca
/]
.
T
h
e
d
ata
s
et
co
n
s
is
ts
o
f
1
5
0
T
1
w
e
ig
h
ted
i
m
ag
e
s
w
it
h
1
m
m
s
lice
t
h
ic
k
n
ess
,
d
if
f
er
en
t
n
o
is
e
le
v
els
(
0
%
,
1
%,
3
%
,
7%
,
9
%)
an
d
d
if
f
er
e
n
t i
n
te
n
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
(
0
%,
2
0
%
,
4
0
%).
Fig
u
r
e
4
.
Qu
er
y
I
m
a
g
e
a
n
d
5
R
elev
a
n
t I
m
ag
e
s
3
.
1
.
Resul
t
o
f
H
is
t
o
g
ra
m
o
f
F
uzzy
O
rient
ed
G
ra
dients
I
n
F
i
g
u
r
e
4
,
t
h
e
a
v
er
ag
e
r
a
n
k
v
er
s
e
s
n
u
m
b
er
o
f
i
m
a
g
e
s
r
etr
iev
ed
i
n
d
if
f
er
e
n
t
n
o
is
e
lev
el
s
o
f
His
to
g
r
a
m
o
f
F
u
zz
y
Or
ien
ted
Gr
ad
ien
ts
ar
e
s
h
o
w
n
.
Fi
g
u
r
e
5
(
a)
s
h
o
w
s
a
v
er
ag
e
r
an
k
i
n
r
etr
iev
in
g
f
ir
s
t
1
5
r
elev
an
t
i
m
a
g
es
w
i
th
0
%
n
o
is
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
t
en
s
it
y
n
o
n
-
u
n
i
f
o
r
m
i
t
y
.
Fi
g
u
r
e
5
(
b
)
s
h
o
w
s
a
v
er
ag
e
r
an
k
o
f
i
m
ag
e
s
w
it
h
1
%
n
o
is
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
ten
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
.
Fi
g
u
r
e
5
(
c)
s
h
o
w
s
a
v
er
ag
e
r
an
k
o
f
i
m
a
g
es
w
it
h
3
%
n
o
i
s
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
te
n
s
it
y
n
o
n
-
u
n
if
o
r
m
it
y
.
Fi
g
u
r
e
5
(
d
)
s
h
o
w
s
a
v
er
ag
e
r
an
k
o
f
i
m
ag
e
s
w
it
h
5
%
n
o
is
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
ten
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
.
Fi
g
u
r
e
5
(
e)
s
h
o
w
s
a
v
er
ag
e
r
an
k
o
f
i
m
a
g
es
w
it
h
7
%
n
o
i
s
e
an
d
0
%,
2
0
%
an
d
4
0
%
i
n
ten
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
An
d
Fig
u
r
e
5
(
f
)
s
h
o
ws
av
er
ag
e
r
an
k
o
f
i
m
a
g
e
s
w
ith
9
%
n
o
is
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
ten
s
i
t
y
n
o
n
-
u
n
if
o
r
m
it
y
.
B
y
a
n
al
y
z
in
g
th
e
s
e
r
esu
lt
s
th
er
e
ca
n
b
e
co
n
cl
u
d
ed
th
at
as i
n
te
n
s
it
y
n
o
n
-
u
n
i
f
o
r
m
i
t
y
in
cr
ea
s
es p
er
f
o
r
m
a
n
ce
d
ec
r
ea
s
es sl
i
g
h
tl
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
8
–
17
14
(
a)
(
b
)
(
c)
(
d
)
(
e)
(
f
)
Fig
u
r
e
5
.
Av
er
ag
e
R
a
n
k
v
er
s
e
s
th
e
N
u
m
b
er
o
f
R
elev
a
n
t I
m
a
g
es
R
etr
iev
ed
w
it
h
Di
f
f
er
en
t
No
is
e
L
e
v
els a
n
d
Dif
f
er
en
t I
n
ten
s
it
y
N
on
-
u
n
i
f
o
r
m
it
y
(
“RF
”)
w
it
h
HF
OF
3
.
2
.
Resul
t
o
f
F
uzzy
L
o
ca
l Bi
na
ry
P
a
t
t
er
ns
I
n
f
i
g
u
r
e
5
,
th
e
av
er
ag
e
r
a
n
k
v
er
s
es
n
u
m
b
er
o
f
i
m
a
g
es
r
etr
i
ev
ed
in
d
i
f
f
er
e
n
t
n
o
i
s
e
lev
e
ls
o
f
Fu
zz
y
L
o
ca
l
B
in
ar
y
P
atter
n
s
ar
e
s
h
o
w
n
.
Fi
g
u
r
e
6
(
a)
s
h
o
w
s
av
er
a
g
e
r
an
k
in
r
etr
iev
i
n
g
f
ir
s
t
1
5
r
e
lev
an
t
i
m
a
g
es
w
it
h
0
%
n
o
is
e
an
d
0
%,
2
0
%
a
n
d
4
0
%
in
te
n
s
i
t
y
n
o
n
-
u
n
i
f
o
r
m
it
y
.
Fig
u
r
e
6
(
b
)
s
h
o
w
s
a
v
er
ag
e
r
an
k
o
f
i
m
a
g
e
s
w
it
h
1
%
n
o
is
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
ten
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
.
Fig
u
r
e
6
(
c)
s
h
o
w
s
av
er
a
g
e
r
an
k
o
f
i
m
a
g
es
w
it
h
3
%
n
o
is
e
an
d
0
%,
2
0
%
a
n
d
4
0
%
in
te
n
s
i
t
y
n
o
n
-
u
n
i
f
o
r
m
it
y
.
Fig
u
r
e
6
(
d
)
s
h
o
w
s
a
v
er
ag
e
r
an
k
o
f
i
m
a
g
e
s
w
it
h
5
%
n
o
is
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
ten
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
.
Fig
u
r
e
6
(
e)
s
h
o
w
s
av
er
a
g
e
r
an
k
o
f
i
m
a
g
es
w
it
h
7
%
n
o
is
e
an
d
0
%,
2
0
%
an
d
4
0
%
in
ten
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
An
d
Fig
u
r
e
6
(
f
)
s
h
o
w
s
av
er
ag
e
r
an
k
o
f
i
m
ag
e
s
w
it
h
9
%
n
o
is
e
a
n
d
0
%,
2
0
%
an
d
4
0
%
in
te
n
s
i
t
y
n
o
n
-
u
n
if
o
r
m
i
t
y
.
B
y
a
n
al
y
zi
n
g
th
e
s
e
r
esu
lt
s
t
h
er
e
ca
n
b
e
co
n
clu
d
ed
th
at
a
s
in
te
n
s
it
y
n
o
n
-
u
n
i
f
o
r
m
it
y
i
n
cr
ea
s
es p
er
f
o
r
m
an
ce
d
ec
r
ea
s
e
s
s
li
g
h
tl
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
B
I
R
o
f B
r
a
in
MR
I
ma
g
es Usin
g
His
to
g
r
a
m
o
f F
u
z
z
y
Ori
en
ted
Gra
d
ien
ts
a
n
d
…
(
A
th
ir
a
TR
)
15
(
a)
(
b
)
(
c)
(
d
)
(
e)
(
f
)
Fig
u
r
e
6
.
Av
er
ag
e
R
a
n
k
v
er
s
e
s
T
h
e
Nu
m
b
er
o
f
R
e
lev
a
n
t I
m
ag
es
R
etr
iev
ed
w
it
h
Di
f
f
er
en
t
No
is
e
L
e
v
els a
n
d
Dif
f
er
en
t I
n
te
n
s
it
y
No
n
-
U
n
i
f
o
r
m
i
t
y
(
“
R
F”)
w
it
h
FL
B
P
3.
3
.
Co
m
pa
ri
s
o
n o
f
H
F
O
G
a
nd
F
L
B
P
I
n
T
ab
le
I
in
d
icate
s
t
h
e
co
m
p
ar
is
o
n
o
f
H
FOG
a
n
d
FL
B
P
f
o
r
th
e
r
etr
ie
v
al
o
f
1
5
r
elev
a
n
t
i
m
a
g
e
s
.
I
t
also
u
s
es
t
h
e
s
a
m
e
d
ataset
u
s
ed
ab
o
v
e.
Fro
m
t
h
e
T
ab
le
1
a
n
d
Fig
u
r
e
7
w
e
ca
n
co
n
clu
d
e
t
h
at
HFOG
o
u
tp
er
f
o
r
m
s
FL
B
P
.
T
ab
le
1
.
C
o
m
p
ar
is
o
n
o
f
HFO
G
an
d
FL
B
P
f
o
r
th
e
R
e
tr
iev
al
o
f
1
5
R
elev
an
t I
m
a
g
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8938
IJ
-
AI
Vo
l.
6
,
No
.
1
,
Ma
r
ch
2
0
1
7
:
8
–
17
16
Fig
u
r
e
7
.
C
o
m
p
ar
is
o
n
o
f
H
FO
G
an
d
FL
B
P
3
.
4
.
Co
m
pa
ri
s
o
n o
f
L
B
P
a
nd
F
L
B
P
Fig
u
r
e
8
s
h
o
w
s
co
m
p
ar
is
o
n
o
f
HF
OG
a
n
d
F
L
B
P
.
Fro
m
t
h
er
e
b
o
th
o
f
t
h
e
m
g
i
v
es
g
o
o
d
r
esu
lt
b
u
t
HFOG
o
u
tp
er
f
o
r
m
s
F
L
B
P
s
lig
h
tl
y
.
Fig
u
r
e
8
.
C
o
m
p
ar
i
s
o
n
o
f
L
B
P
an
d
FL
B
P
4.
CO
NCLU
SI
O
N
Her
e
p
r
o
p
o
s
es
t
w
o
m
eth
o
d
s
,
His
to
g
r
a
m
o
f
F
u
zz
y
Or
ien
t
ed
Gr
ad
ien
ts
a
n
d
F
u
zz
y
L
o
c
al
B
in
ar
y
P
atter
n
f
o
r
B
r
ain
MR
I
m
a
g
e
r
etr
iev
al.
I
t
is
p
r
o
v
en
th
at
HO
G
is
a
g
o
o
d
d
escr
ip
to
r
f
o
r
b
r
ain
I
m
ag
e
r
etr
iev
al.
I
n
HOG,
lar
g
er
th
e
n
u
m
b
er
i
s
m
o
r
e
ac
c
u
r
ate
th
e
d
escr
ip
to
r
is
.
B
u
t
th
e
i
m
p
ac
t
o
f
th
e
b
in
n
u
m
b
er
af
f
ec
ts
t
h
e
d
is
tr
ib
u
tio
n
o
f
th
e
h
is
to
g
r
a
m
.
So
,
h
er
e
co
m
e
to
t
h
e
co
n
ce
p
t
o
f
His
to
g
r
a
m
o
f
Fu
zz
y
Or
ie
n
ted
Gr
ad
ien
ts
.
I
n
HFOG
ea
c
h
p
i
x
el
b
elo
n
g
i
n
g
m
o
r
e
th
a
n
o
n
e
b
in
at
t
h
e
s
a
m
e
ti
m
e
w
i
th
d
if
f
e
r
en
t
d
eg
r
ee
s
.
Si
m
ilar
l
y
f
u
zz
if
ica
tio
n
a
llo
w
s
a
F
u
zz
y
L
o
ca
l
B
in
ar
y
P
atter
n
(
F
L
B
P
)
to
co
n
tr
ib
u
te
to
m
o
r
e
th
a
n
a
s
i
n
g
le
b
in
in
t
h
e
d
is
tr
ib
u
tio
n
o
f
th
e
L
B
P
v
alu
e
s
u
s
ed
as a
f
ea
t
u
r
e
v
ec
to
r
.
RE
F
E
R
E
NC
E
S
[1
]
S
m
e
u
ld
e
rs,
M
W
o
rrin
g
,
S
S
a
n
ti
n
i,
A
G
u
p
ta,
R
Ja
in
,
Co
n
ten
t
b
a
se
d
im
a
g
e
re
tri
e
v
a
l
a
t
th
e
e
n
d
o
f
th
e
e
a
rly
y
e
a
rs.
Pa
tt
e
rn
A
n
a
lys
is
a
n
d
M
a
c
h
i
n
e
In
t
e
ll
ig
e
n
c
e
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
.
2
0
0
0
;
2
2
:
1
3
4
9
-
1
3
8
0
.
[2
]
Zu
k
u
a
n
,
W
EI,
e
t
a
l.
A
n
Eff
icie
n
t
Co
n
te
n
t
Ba
se
d
Im
a
g
e
Re
tri
e
v
a
l
S
c
h
e
m
e
.
T
EL
KOM
NIKA
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
.
2
0
1
3
;
1
1
(
1
1
)
:
6
9
8
6
-
6
9
9
1
.
[3
]
M
ü
ll
e
r,
He
n
n
i
n
g
,
T
h
o
m
a
s
M
De
se
rn
o
.
Co
n
ten
t
-
b
a
se
d
m
e
d
ica
l
im
a
g
e
re
tri
e
v
a
l.
Bio
m
e
d
ica
l
I
m
a
g
e
P
ro
c
e
ss
in
g
.
S
p
rin
g
e
r B
e
rlin
He
i
d
e
lb
e
rg
.
2
0
1
1
:
4
7
1
-
4
9
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
AI
I
SS
N:
2252
-
8938
C
B
I
R
o
f B
r
a
in
MR
I
ma
g
es Usin
g
His
to
g
r
a
m
o
f F
u
z
z
y
Ori
en
ted
Gra
d
ien
ts
a
n
d
…
(
A
th
ir
a
TR
)
17
[4
]
L
e
h
m
a
n
n
,
T
h
o
m
a
s
M
,
e
t
a
l.
Co
n
ten
t
-
b
a
se
d
im
a
g
e
re
tri
e
v
a
l
in
m
e
d
ica
l
a
p
p
li
c
a
ti
o
n
s:
a
n
o
v
e
l
m
u
lt
istep
a
p
p
ro
a
c
h
.
El
e
c
tro
n
ic Im
a
g
in
g
.
In
tern
a
ti
o
n
a
l
S
o
c
iety
f
o
r
Op
ti
c
s an
d
P
h
o
to
n
ics
,
1
9
9
9
.
[5
]
X
ia,
L
in
g
,
e
t
a
l.
M
e
d
ica
l
I
m
a
g
e
Re
tri
e
v
a
l
B
a
se
d
o
n
S
h
a
p
e
F
e
a
tu
r
e
s
in
DCT
Do
m
a
in
.
T
EL
KOM
NIKA
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
2
0
1
4
;
1
2
(2
)
:
1
1
1
6
-
1
1
2
4
.
[6
]
F
e
li
p
e
,
Jo
a
q
u
im
Ce
z
a
r,
Ag
m
a
JM
T
ra
in
a
,
Ca
e
tan
o
T
ra
in
a
Jr.
Re
tri
e
v
a
l
b
y
Co
n
ten
t
o
f
M
e
d
ica
l
I
m
a
g
e
s
U
sin
g
T
e
x
tu
re
f
o
r
T
issu
e
I
d
e
n
ti
f
ica
ti
o
n
.
Co
m
p
u
ter
-
Ba
se
d
M
e
d
ica
l
S
y
ste
m
s,
2
0
0
3
.
Pro
c
e
e
d
in
g
s.
1
6
t
h
IEE
E
S
y
mp
o
siu
m
.
IEE
E,
2
0
0
3
.
[7
]
Da
lal,
Na
v
n
e
e
t,
Bil
l
T
rig
g
s.
Histo
g
ra
m
s
o
f
Orie
n
ted
G
r
a
d
ien
ts
f
o
r
Hu
m
a
n
D
e
tec
ti
o
n
.
Co
m
p
u
ter
Vi
sio
n
a
n
d
P
a
tt
e
r
n
Re
c
o
g
n
it
io
n
,
2
0
0
5
.
C
VP
R
2
0
0
5
.
I
EE
E
Co
m
p
u
ter
S
o
c
iety
Co
n
fer
e
n
c
e
o
n
.
I
EE
E
,
2
0
0
5
;
1
.
[8
]
Ov
e
re
tt
,
G
a
r
y
,
L
a
rs
P
e
ters
so
n
.
L
a
r
g
e
S
c
a
le
S
ig
n
De
tec
ti
o
n
Us
in
g
HO
G
F
eat
u
re
V
a
rian
t
s
.
In
te
ll
ig
e
n
t
Ve
h
icle
s
S
y
mp
o
si
u
m (
IV)
,
IEE
E
,
2
0
1
1
.
[9
]
S
a
lh
i,
A
b
d
e
l
Ilah
,
M
u
sta
p
h
a
Ka
rd
o
u
c
h
i
,
Na
b
il
Be
lac
e
l.
Hist
o
g
ra
m
s
o
f
F
u
z
z
y
Orie
n
ted
G
r
a
d
ien
ts
f
o
r
F
a
c
e
R
e
c
o
g
n
it
io
n
.
Co
m
p
u
ter
A
p
p
l
i
c
a
ti
o
n
s T
e
c
h
n
o
l
o
g
y
(
ICCAT
),
I
n
ter
n
a
ti
o
n
a
l
C
o
n
fer
e
n
c
e
o
n
.
IE
EE
,
2
0
1
3
.
[1
0
]
Oja
la,
T
i
m
o
,
M
a
tt
i
P
ietik
ä
in
e
n
,
T
o
p
i
M
ä
e
n
p
ä
ä
.
M
u
lt
ires
o
l
u
ti
o
n
g
ra
y
-
sc
a
le
a
n
d
ro
tatio
n
i
n
v
a
rian
t
tex
tu
re
c
las
si
f
ica
ti
o
n
w
it
h
lo
c
a
l
b
i
n
a
ry
p
a
tt
e
rn
s.
Pa
tt
e
rn
An
a
lys
is
a
n
d
M
a
c
h
in
e
In
tell
ig
e
n
c
e
,
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
2
0
0
2
;
24
(
7
):
9
7
1
-
9
8
7
.
[1
1
]
Bra
h
n
a
m
,
S
h
e
ry
l,
e
t
a
l.
L
o
c
a
l
b
in
a
r
y
p
a
tt
e
rn
s: n
e
w
v
a
rian
ts
a
n
d
a
p
p
li
c
a
ti
o
n
s
.
S
p
rin
g
e
r B
e
rlin
He
i
d
e
lb
e
rg
,
2
0
1
4
.
[1
2
]
Ia
k
o
v
id
is,
Dim
it
ris
K,
E
y
stra
ti
o
s
G
Ke
ra
m
id
a
s,
Di
m
it
ris
M
a
ro
u
li
s.
F
u
z
z
y
lo
c
a
l
b
in
a
ry
p
a
tt
e
rn
s
f
o
r
u
lt
ra
so
u
n
d
tex
tu
re
c
h
a
ra
c
teriz
a
ti
o
n
.
Im
a
g
e
A
n
a
l
y
sis
a
n
d
Re
c
o
g
n
it
i
o
n
.
S
p
rin
g
e
r B
e
rlin
He
id
e
lb
e
rg
,
2
0
0
8
:
7
5
0
-
7
5
9
.
BI
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
M
s.
A
th
ira
T
R
,
P
G
S
c
h
o
lar
in
A
d
i
S
h
a
n
k
a
ra
In
stit
u
te
o
f
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
sp
e
c
ializa
ti
o
n
in
C
S
E
u
n
d
e
r
M
G
U.
B.
T
e
c
h
.
d
e
g
re
e
in
Co
m
p
u
t
e
r
S
c
ien
c
e
En
g
in
e
e
rin
g
f
ro
m
M
G
U,
In
d
ia,
i
n
2
0
1
3
.
Dr.
A
b
ra
h
a
m
V
a
rg
h
e
se
,
p
re
se
n
tl
y
w
o
rk
in
g
a
s
a
n
A
ss
o
c
iate
p
ro
f
e
ss
o
r
in
De
p
a
rtm
e
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
,
A
d
i
S
h
a
n
k
a
ra
In
stit
u
te
o
f
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
.
Co
m
p
lete
d
P
h
D
f
ro
m
A
n
n
a
u
n
iv
e
rsit
y
Ch
e
n
n
a
i
a
n
d
M
.
T
e
c
h
(CS
E)
f
ro
m
Na
t
io
n
a
l
in
stit
u
te
o
f
T
e
c
h
n
o
lo
g
y
Ka
rn
a
tak
a
(NI
T
K),
S
u
ra
th
k
a
l
.
He
h
a
s
ten
y
e
a
r
e
x
p
e
rien
c
e
in
tea
c
h
in
g
a
n
d
f
iv
e
y
e
a
r
e
x
p
e
rien
c
e
in
re
se
a
rc
h
.
Evaluation Warning : The document was created with Spire.PDF for Python.