Specifications
Image Statistics Functions
12-3
12
(1) the norm of
srcImageA
pixel values,
||
a
||
(2) the norm of differences of the source images’ pixel values,
||
a - b
||
(3)
the relative error
||
a - b
||
/
||
b
||
(see formulas below).
Let a ={a
k
}andb ={b
k
} be vectors containing pixel values of
srcImageA
and
srcImageB
, respectively (all channels are used except alpha channel).
(1)Ifthe
srcImageB
pointer is NULL, the function returns the norm of
srcImageA
pixel values:
||
a
||
L
1
=
Σ
k
|
a
k
|
for
normType
= IPL_L1
||
a
||
L
2
= (
Σ
k
|
a
k
|
2
)
1/2
for
normType
= IPL_L2
||
a
||
C
=max
k
|
a
k
|
for
normType
= IPL_C.
(2)Ifthe
srcImageB
pointer is not NULL, the function returns the norm of
differences of
srcImageA
and
srcImageB
pixel values:
||
a - b
||
L
1
=
Σ
k
|
a
k
- b
k
|
for
normType
= IPL_L1
||
a - b
||
L
2
= (
Σ
k
|
a
k
- b
k
|
2
)
1/2
for
normType
= IPL_L2
||
a - b
||
C
=max
k
|
a
k
- b
k
|
for
normType
= IPL_C.
(3) If
normType
is IPL_RELATIVEC, IPL_RELATIVEL1,or
IPL_RELATIVEL2,the
srcImageB
pointer must not be NULL.
The function first computes the norm of differences, as defined in (2). Then
this norm is divided by the norm of b, and the function returns the relative
error
||
a - b
||
/
||
b
||.
Return Value
The computed norm or relative error in double floating-point format.