Biomedical Engineering Reference
In-Depth Information
50
45
40
35
30
25
20
15
10
5
0
0
10
20
30
40
50
60
70
Fig. 12
Average horizontal summation
P
L
k:f .x C k/
k
D
L
d.x/ D
;
P
L
k
2
k
D
L
where
x
is point where the general difference is computed,
d(x)
is the value of the
general difference in point
x
,
L
is half of the width of the weighting window,
k
is
point of the surroundings and it's weight.
Then, we minimum find zero crossing point in first difference and in second
difference that of this curve that represented maximum of the original curve - global
extreme. To gain high precision we interpolate surroundings of this extreme. Result
is the horizontal position of the 1st PI.
For convergence analysis we subtract 1st PI and x
j
, that represents distance
between 1st PI and COL and show us time demanding process of convergence.
First step of accommodation analysis is to remove the 1st PI from thresholded
ROI by a new partial thresholding. New threshold is set in 70% of dynamic
range and it eliminates higher values of brightness that represent 1st PI. Then
average horizontal summation (the same as in convergence analysis but in horizontal
direction) is presented in Fig.
12
. This is the horizontal summation in pixel rows
devided by number of nonzero pixels in the same row. By fitting middle part of this
curve we get polynomial of the first order:
y
D a
0
C a
1
x:
