Photon Focus MV1-D1312C CameraLink Series, User Manual

Page: 51 / 99

4.7 Digital Gain and Offset
Gain x1, x2, x4 and x8 are digital amplifications, which means that the digital image data are
multiplied in the camera module by a factor 1, 2, 4 or 8, respectively. It is implemented as a
binary shift of the image data, which means that there will be missing codes in the output
image as the LSB’s of the gray values are set to ’0’. E.g. for gain x2, the output value is shifted
by 1 and bit 0 is set to ’0’.
A user-defined value can be subtracted from the gray value in the digital offset block. This
feature is not available in Gain x1 mode. If digital gain is applied and if the brightness of the
image is too big then the output image might be saturated. By subtracting an offset from the
input of the gain block it is possible to avoid the saturation.
4.8 Fine Gain
The MV1-D1312C camera series have to types of fine gains included. An image overall fine gain
and RGB channel fine gain.
The image fine gain can be used to adjust the brightness of the whole image.
The RGB channel fine gain is used to calibrate the white balance in an image, which has to be
set according to the current lighting condition.
The RGB gain is multiplied by the image overall fine gain and digital gain.
4.9 Channel Colour Level Transformation (LUT)
Channel colour level transformation is remapping of the colour level values of an input image
to new values. The look-up table (LUT) is used to convert the channel colour value of each pixel
in an image into another channel colour value. It is typically used to implement a transfer
curve for contrast expansion. The camera performs a 12-to-8-bit mapping, so that 4096 input
channel colour levels can be mapped to 256 output channel colour levels. The use of the three
available modes is explained in the next sections. Two LUT and a Region-LUT feature are
available in the MV1-D1312C camera series (see Section 4.9.4).
The output channel colour level resolution of the look-up table (independent of
gain, gamma or user-definded mode) is always 8 bit.
There are 2 predefined functions, which generate a look-up table and transfer it
to the camera. For other transfer functions the user can define his own LUT file.
Some commonly used transfer curves are shown in Fig. 4.33. Line a denotes a negative or
inverse transformation, line b enhances the image contrast between colour channel values x0
and x1. Line c shows brightness thresholding and the result is an image with only black and
white levels. and line d applies a gamma correction (see also Section 4.9.2).
4.9.1 Gain
The ’Gain’ mode performs a digital, linear amplification with clamping (see Fig. 4.34). It is
configurable in the range from 1.0 to 4.0 (e.g. 1.234).
4.7 Digital Gain and Offset 49