Image Processing Reference
In-Depth Information
0
142
1
123
2
130
3
135
4
138
5
135
6
130
7
123
8
121
9
125
0
1
2
3
4
5
6
7
8
9
10
11
12
151
0
0
0
0
0
0
0
0
0
157
0
0
0
0
0
0
0
0
0
160
0
0
0
0
0
0
0
0
0
161
0
0
0
0
0
1
0
0
0
floor(sc_smooth-smoothed) =
162
0
0
0
0
0
0
0
0
0
162
0
0
0
0
0
0
0
0
0
161
0
1
0
0
0
0
0
0
0
157
0
0
0
0
1
0
0
0
0
157
0
0
0
0
0
0
0
0
0
157
0
0
0
0
0
0
0
0
0
159
0
0
0
0
1
1
0
0
0
159
0
0
0
0
0
0
0
0
0
which shows that the difference is in the borders, the small differences in pixels' values are
due to numerical considerations.
In image processing, we often use a
Gaussian
smoothing filter which can give a better
smoothing performance than direct averaging. Here the template coefficients are set according
to the Gaussian distribution which for a two-dimensional distribution controlled by a
variance σ
2
is, for a template size defined by
winsize:
Gaussian_template(winsize,
σ
):= sum
←
0
winsize)
2
centre
←
floor
for y
∈
0..winsize-1
for x
∈
0..winsize-1
2
2
-[(y-centre) +(x-centre) ]
2
⋅σ
2
template
y,x
←
e
sum
←
sum+template
y,x
template
sum
So let's have a peep at the normalised template we obtain:
0.003
0.013
0.022
0.013
0.003
0.013
0.06
0.098
0.06
0.013
0.022
0.098
0.162
0.098
0.022
Gaussian_template(5,1)=
0.013
0.06
0.098
0.06
0.013
0.003
0.013
0.022
0.013
0.003
