The DGV image receiver employs a pair of thermo-electrically cooled CCD cameras with a dynamic range of 12 bits, which view the obervation area through a common objective lens (right). Due to the limited back-focal length of the objective lens a relay lens pair projects the first image onto the CCD sensors. A non-polarizing 50%-50% beam splitter following the relay lens pair redirects part of the collected light onto the reference CCD while the remaining light passes through the iodine absorption cell before reaching the signal CCD. Although the relay lens degrades the image resolution (MTF), the major advantage of this setup is that a change of focus or the replacement of the objective lens does not require a re-calibration of the two CCD sensors with respect to each other. Also the use of a flexible endoscope for viewing flows with difficult optical access is possible. The more common dual objective lens configuration as described in literature requires a complete calibration of the system each time a lens is altered. However a higher light sensitivity and better image resolution are the major advantages in that case.