Image Processing Reference
In-Depth Information
trun_med(p,wsze):= newpic
←
zero(p)
ha floor
wsze
2
for x
∈
ha..cols(p)-ha-1
for y
∈
ha..rows(p)-ha-1
win
←
submatric(p, y-ha, y+ha, x-ha, x+ha)
med
←
median(win)
ave
←
mean(win)
upper
←
2.med-min(win)
lower
←
2.med-max(win)
cc
←
0
for i
∈
0..wsze-1
for j
∈
O..wsze-1
trun
cc
←
win
j,i
if(win
j,i
<upper)·(med<ave)
trun
cc
←
win
j
,
i
if(win
j,i
>lower)·(med>ave)
cc
←
cc+1
newpic
y,x
←
median(turn)
newpic
Code 3.14
The truncated median operator
because the scanner is usually external to the body. The noise is actually multiplicative
Rayleigh noise for which the mode is the optimal estimate. This noise obscures the artery
which appears in cross-section in Figure
3.24
(a); the artery is basically elliptical in shape.
The action of the 9 × 9 truncated median operator, Figure
3.24
(b) is to remove noise whilst
retaining feature boundaries whilst a larger operator shows better effect, Figure
3.24
(c).
(a) Part of ultrasound image
(b) 9
9 operator
(c) 13
13 operator
Figure 3
.
24
Applying truncated median filtering
Close examination of the result of the truncated median filter is that a selection of
boundaries is preserved which is not readily apparent in the original ultrasound image.
