ALLIED Vision Technologies Pike F-1600 - Manual (Page 190)

Description of the data path
PIKE Technical Manual V5.1.2
190
High SNR mode (High Signal Noise Ratio)
In this mode the camera grabs and averages a set number of images and outputs
one image with the same bit depth and the same brightness. This means that the
camera will output an 8-bit averaged image when an 8-bit image format is
selected (although the internal calculations are done with 14 bit).
Because of the fact that normally uncorrelated (photon-, amplifier-) noise
dominates over correlated noise (fixed pattern noise), adding two images will
double (6 dB) the gray levels but only increase the noise levels by
(3 dB).
This enhances both the dynamic range as well as the signal-to-noise ratio.
Consequently adding 256 8-bit images will lead to a potential signal-to-noise
enhancement of 24 dB or a resulting bit depth of 16 bit.
Note

Configuration
To configure this feature in an advanced register: See Table
175: Advanced register: High Signal Noise Ratio (HSNR) on
page 354.
Note

• The camera must be idle to toggle this feature on/off.
Idle means: no image acquisition, no trigger.
• Set grab count and activation of HighSNR in one single
write access.
Note

• The averaged image is output at a lower frame rate
roughly equivalent to fps_old/N, where N is the number
of images averaged. In fact, due to camera internal con-
ditions, and according to which format and mode set-
tings are in use, it can vary slightly to be closer some-
times to 1/ ((N/fps_old) + T_shutter). It's impractical to
express in a formula or tables, across all camera models
and modes. But these notes should be sufficient to help
each user determine that the camera behaves as
described.
• The potential SNR enhancement may be lower when using
more than 8-bit original bit depth.
• Select 16-bit image format in order to take advantage of
the full potential SNR and DNR (DyNamic Range)
enhancements.
• For 8-bit video modes, the internal HSNR calculations are
done with 14 bit.
2