Eclipse EC-11-xxx40, Руководство пользователя

Страница: 42 / 56

42
EC-11User’s Manual
03-32-00418-05 DALSA Corp.
To perform dark calibration:
1. Stop all light from entering the camera.
2. Issue the command
correction_calibrate_fpn
. The camera will respond with
OK> if no error occurs. V
input
needs to be between 1DN and 127DN, otherwise the
correction will be stopped. Use the
get_line
command to determine input level.
Choose low FPN coefficient values to achieve optimum FPN calibration performance.
3. Dark calibration automatically calibrates FPN coefficients and digital offset. This
procedure needs to be repeated for forward and reverse directions, if direction control
is used.
4. After the calibration is complete, you can save these settings, and the PRNU
coefficients, to non-volatile memory so they will be remembered after power-down.
To do so, issue the command
write_pixel_coeffs
.
To perform a dark calibration with an analog offset value:
Use the
correction_calibrate_fpn value
command. Where value is the analog
offset DN. This command sets the analog offset first to value before calculating the FPN
coefficients range from 1 to 100. Use this command in calibrated mode only. If value is left
blank, current offset values are used.
White Light Calibration
White light calibration is more complex than dark calibration because the camera
attempts to create a flat white image. This calibration corrects PRNU effects as well as
non-uniform lighting and lens vignetting affects.
White light calibration requires a clean, white reference. The quality of this reference is
important for proper calibration. White paper is often not sufficient because the grain in
the white paper will distort the correction. A more uniform source such as white plastic
will lead to better balancing.
The camera adjusts to the brightest pixel plus largest FPN pixel subtraction value.
There are several restrictions that must be met in order for the calibration to succeed:
1. The camera is sufficiently sensitive to detect 60 Hz ambient light flicker which may
affect camera performance and calibration results.
2. The light level must be set so that input to the digital processing chain is between
128DN and 254DN, otherwise calibration will stop. Use the
get_line
command to
determine input level.
3. The variance in light intensity across the target must not be more than 2 to 1. Since the
maximum per-pixel digital gain is 2x, the camera will not be able to compensate for
extremely non-uniform light.
These restrictions are all tested within the calibration algorithm and the camera will
report an error if any of these conditions could not be met. In this case, the calibration will
stop.
To perform a white light calibration:
1. Place a white reference in front of the camera.
Note:
If your
illumination or white
reference does not extend
the full field of view of the
camera, calibration will not
be successful. The camera
will send a warning.