Falcon 4M Camera Manual
DALSA 03-032-20044-01
43
3)
When 6.25% of pixels from a single row within the region of interest are clipped, flat
field correction results may be inaccurate.
4)
Correction results are valid only for the current analog offset values. If you change
this value, it is recommended that you recalculate your coefficients.
Let’s go through a flat field calibration example:
1)
The camera is placed in
sem 2
(no other exposure mode will allow FFC calibration)
2)
Settings such as frame rate, exposure time, etc. are set as close as possible to the
actual operating conditions. Set digital gain to X1 (
ssg 0 4096
) and background
subtract to 0 (
ssb 0 0
) as these are the defaults during FFC calibration.
3)
Place the camera in the dark and send CCF, this performs the FPN correction and
automatically save the FPN coefficients to non-volatile memory
4)
Set
epc 1 0
, which enables the FPN correction and verify the signal output is close
to 0 DN. Leave
epc 1 0
for the next step since the
cpa
target assumes there is no
FPN. This is important on the 4M60/30 due to the large dark offset values.
5)
Illuminate the sensor, such that with EPC 1 0, it reaches 50-70% saturation.
6)
Send
cpa 2 T
where T is typically 1.3X the average output level. This is important
since if the target is too low (<1.1X), then some pixels may not be able to reach full
swing (1023 DN) due to corrections applied by the camera.
Here is the factory calibration procedure for Snapshot Mode 1 (
efd 1
):
1)
The camera is placed in
sem 2
,
sot 320
,
clm 16
,
efd 1
,
snd 1
, full window,
ssg
0 4096
,
ssb 0 0
,
sao 0 0
,
ssf 55
,
set 2000
. This last part is important,
ssf
55
and
set 2000
assures that the camera is in non-concurrent mode. In non-
concurrent mode, readout and integration do not overlap thus eliminating some
residual artifacts associated with concurrent mode.
2)
The camera is placed in the dark and
ccf
is run
3)
With
epc 1 0
the sensor is illuminated (Light Source: Broadband Quartz Halogen,
3250K, with a 750 nm cutoff filter) with a light level of 22.8 W/cm
2
. This ensures each
camera will have the same responsivity since the light level and target value are
always the same. Typical output levels for the camera at this light level are 650.
4)
The sensor window at this point has been cleaned thoroughly such that there are no
significant blemishes present.
5)
Send
cpa 2 840
. Typically this yields an average PRNU coefficient of about 1.3X.
