Image Processing Reference
In-Depth Information
b. High-pass
filters (HPF)
2
4
3
5
2
4
3
5
0
10
1
21
h
(
n, m
) ¼
15
1
,
h
(
n, m
) ¼
25
2
,
0
10
1
21
2
4
3
5
2
4
3
5
1
2
1
1
1
1
1
7
h
(
n, m
) ¼
2 9
2
,
h
(
n, m
) ¼
19
1
(
2
:
83
)
1
2
1
1
1
1
The FIR
filtering of an image is performed one pixel at a time. The center
lter
weight is placed on the pixel to be
filter tap is multiplied by its
corresponding pixel value, and the results are summed together. As an example,
consider
filtered. Each
filtering the pixel f
(
5, 2
)
in the following 8
8 block of the LENA image
2
3
139
144
149
153
155
155
155
155
4
5
144
151
153
156
159
156
156
156
150
155
160
163
158
156
156
156
159
161
162
160
159
159
158
159
---------------------
f
(
n, m
) ¼
(
2
:
84
)
159
160
161
162
162
155
155
155
161
161
161
161
160
157
157
157
162
162
161
163
162
157
157
157
---------------------
162
162
161
161
163
158
158
158
Using the 3
3 FIR
lter
2
4
3
5
121
242
121
1
16
h
(
n, m
) ¼
(
2
:
85
)
the output pixel g
(
5, 2
)
is computed as
1
16
(
g
(
5, 2
) ¼
4
161
þ
2
161
þ
2
161
þ
2
161
þ
2
161
þ
1
162
þ
1
162
þ
1
160
þ
1
163
) ¼
161
(
2
:
86
)
FIR
filters are used in many image processing tasks such as smoothing, sharpening,
noise removal, and edge detection [15
-
17]. Examples of
filters are
LPF: If the filter weights are all positive, the filtering operation creates a
smoother (softer) image that consists mostly of low frequencies. These
types of
filters are referred to as LPF. LPF are used as anti-aliasing
filters
or prior to subsampling in image decimation. The larger the
filter, and the
flatter the weights, the smoother the resulting image.
Search WWH ::
Custom Search