OPERATING AND USER MANUAL QUARTZ series CXP
37
Enable / Disable defect pixel correction and diagnosis
6.10.1.
The defect pixel correction is enabled or disabled with the DefectPixelCorrectionEnable register.
As for checking the defect pixel list on top level, some possibilities are implemented:
The total number of defect pixels in the list can be queried with the DefectPixelTotal parameterThe
•
position of a specific pixel in the list can be read out by selecting that pixel with DefectPixelSelect
and querying DefectPixelReadX and DefectPixelReadY.
The coordinates are with respect to the left-top of the image (1,1).
The positions of the defect pixels in the list can be visualized using the DefectPixelTestMode in
•
various ways:
- WIth MarkDefectsWhiteOnVideo, the defect pixels are displayed as white spots in the image
provided by normal operation.
- With MarkDefectsBlackOnVideo, the defect pixels are displayed as black spots in the image
provided by normal operation.
- With ShowDefectsAsWhiteOnBlackBackground, the defect pixels are displayed as white on a
black image.
Modifying the defect pixel list
6.10.2.
Another defect pixel may be added to the list by first setting its coordinates in the camera with
DefectPixelWriteX and DefectPixelWriteY. The actual addition to the list is then by executing the
DefectPixelAdd command. The coordinatex X and Y are with respect to the left-top of the image (1,1).
The defect pixel with coordinates DefectPixelWriteX and DefectPixelWriteY is removed from
the list with the DefectPixelRemove command. Once the combination DefectPixelWriteX and
DefectPixelWriteY is not in the list, the DefectPixelRemove command is ignored.
The entire defect pixel list can be erased with the DefectPixelClearAll command.
At all times, it is possible to restore the factory defect pixel list, with the DefectPixelFactoryRestore
command. Defect pixels that have been added to the list in a later stage, are lost.
Flat Field calibrations and corrections (ICategory: Calibration)
6.11
The pixels of the CMOS sensor show offset and gain variations in their response to light.
In the Quartz cameras these deviations are per-pixel corrected for.
The offset correction is referred to as dark field correction.
•
The gain correction is referred to as bright field correction. It comes in two flavours, local and
•
global.
In local flat field gain correction, a particular pixel is ‘flat fielded’ with respect to it’s near
environment.
In global flat field gain correction, the entire image is ‘flat fielded’ towards the same response (this
can be used to compensate for e.g. shading due to optics).
The gain correction is capable of a maximum gain value of 4x. If the illumination or optics shading
would require higher gains for certain pixels, the calibration will still be completed with the max
gain truncated at 4x. If this situation occurs, a RatioMaximumOverMinimumError is flagged.
The offset and local gain flat field corrections are pre-calibrated in the factory. The user is able to
recalibrate in the field. Calibrations may take up to several seconds.
Up to 4 gain calibrations may be saved for use.
The commands related to flat field calibrations and corrections are listed right below. In the
subsections afterwards they are clarified.
