Cheetah Python Camera with USB3 Interface
–
User Manual
December 5, 2017
Page
73
of
78
Rev 1.0
5.13.2 User Defined LUT
You can define any 12-bit to 12-bit transformation as a user Look-up Table (LUT) and
upload it to the camera using the configuration utility software. You can specify a transfer
function to match the camera’s dynamic range to the scene’s dynami
c range. There are no
limitations to the profile of the function. The LUT must include all possible input values (0
to 4095).
Input signal
O
u
tp
u
t
s
ig
n
a
l
Original TF
Modified TF
Figure 46: Custom LUT.
5.14 Defective Pixel Correction
A CMOS imager is composed of a two-dimensional array of light sensitive pixels. In
general, the majority of the pixels have similar sensitivity. However, some pixels deviate
from the average pixel sensitivity
and are called “defective pixels.”
In most cases,
defective pixels are responsive to light, and rarely is a pixel totally dark or totally bright.
There are two major types of pixel defects: defective and hot.
1.
Defective.
These are pixels whose sensitivity deviates due to fluctuations in the
CMOS manufacturing process and materials. At the factory, final testing
identifies and corrects up to 1024 defective pixels using defective pixel
correction. Two types of defective pixels are possible:
a.
Dark
- a pixel whose sensitivity is lower than the sensitivity of the adjacent
pixels.
b.
Bright
- a pixel whose sensitivity is higher than the sensitivity of the adjacent
pixels.
2.
Hot.
These are pixels that in normal camera operation behave as normal pixels
(sensitivity equal to one of the adjacent pixels), but during long exposures or at
elevated temperatures, the pixel becomes much, much brighter than the
average of the pixels surrounding it. In some cases, the pixel becomes so bright
that the pixel saturates. Final camera testing at the factory identifies and
automatically corrects up to 8192 hot pixels.
