Humanoid-Eye Imaging System Model with Ability of Resolving Power Computing - Advances in Image and Graphics Technologies - page 100

Table 2. Predicted static MDD and results for the effect of ʳ ( d 0 = 1800mm)

10 40 80

ʳ °

4.06 5.22 23.04

 s / mm

image

Table 3. Predicted static MDD and results for the effect of ʳ ( d 0 = 3000mm)

10 40 80

ʳ °

6.77 8.70 38.39

 s / mm

image

Table 4. Predicted static MDD and results for the effect of Ψ

ˈ °

+ 10 + 20

0

 s / mm 8.00 7.76 7.06

image

4.2 Discussions

In the static part, there are 27 out of 35 (approximately 80%) images strictly satisfying
the definition of just distinguishable situation defined in section 4.2. For the rest 20
percent, the error is only one pixel. Allowing for the fact that CCD samples image
discretely to cause an error of a pixel for two LEDs (as discussed below), we treat
these 20 percent's image falling within the error threshold and regard the prediction
made by MDD formula as congruous with the whole static results. In the dynamic
part, 5 out of 13 images strictly match the just distinguishable situation and for the
rest images, 7 out of 8 fall in the error threshold of one pixel. We consider the dynam-
ic results agreeable to the MDD calculation with only one image's violation.
The main source of error in the experiment comes from discretely imaging of
CCD. Rather than an ideal point-photo source, the structure of LED is actually a light-
emitting plane and other factors such as diffraction and diffusion also contribute to the