47
Bias correction file format
The bias correction file contains an image to be subtracted from the frame sent by the sensor.
The bias correction allows to counterbalance the difference of offsets between the different pixels.
Bias correction file geometry must match the actual sensor geometry and must contain pixels values
in the range [0-16383].
Pixels must be stored in left-to-right/top-to bottom order (the same order than the pixels in a
received frame from the sensor). Each pixel value is stored on 16-bits, as for frames received from
the sensor.
Flat correction file format
The flat correction file contains an array of fixed-point real values that are used to counterbalance the
difference of integration ramp slopes between the different pixels.
Flat correction file geometry must match the actual sensor geometry. Values are stored in left-to-
right/top-to-bottom order (the same order than the pixels in a received frame from the sensor).
When flat correction is active, pixel values read from the sensor are multiplied by the corresponding
value in the flat correction file.
Values are stored using fixed-point representation.
Bits 15-13 are used to store the integral part of the multiplication factor.
Bits 12-0 are used to store the decimal part of the multiplication factor.
Examples of values:
-
00100000000000000b / 0x02000000 is the representation of value 1.0
-
01000000000000000b / 0x04000000 is the representation of value 2.0
-
00010000000000000b / 0x01000000 is the representation of value 0.5
-
00110000000000000b / 0x01000000 is the representation of value 1.5
Pixel value offset
It is possible to add an offset to the pixel value.
It can be used to keep the pixel values strictly positive. Indeed, the camera pixel format is 16 bits signed
but some acquisition software cannot manage signed pixels. In this case, adding an offset can be useful
to avoid negative pixels.
To add an offset, please use the command: “
set aduoffset <aduOffsetValue>
”
You can read the currently applied offset using the command: “
aduoffset
”.
Note: It is possible to have negative pixels value when the bias correction is used.
Dark optimization
A parameter optimizes the dark current.
There is a tradeoff to do between the readout noise and the dark current, with the command:
“
set darkoptimlevel [0-100]
”
