Princeton 4300 Projector User Manual


 
50 PI-SCX System Manual Version 2.E
50
The upper left drawing in Figure 19
represents a CCD after exposure but
before the beginning of readout. The
capital letters represent different
amounts of charge, including both
signal and dark charge. This section
explains readout at full resolution,
where every pixel is digitized
separately.
Readout of the CCD begins with the
simultaneous shifting of all pixels
one row toward the “shift register,”
in this case the row at the top. The
shift register is a single line of
pixels along one edge of the CCD,
not sensitive to light and used for
readout only. Typically the shift
register pixels hold twice as much
charge as the pixels in the imaging
area of the CCD.
After the first row is moved into the
shift register, the charge now in the
shift register is shifted toward the
Figure 19. Full Frame at Full Resolution
output node, located at one end of the shift register. As each value is “emptied” into the
output it is digitized. Only after all pixels in the first row are digitized is the second row
moved into the shift register. The order of shifting in our example is therefore A1, B1,
C1, D1, A2, B2, C2, D2, A3 ....
After charge is shifted out of each pixel the remaining charge is zero, meaning that the
array is immediately ready for the next exposure.
A subsection of the CCD can be read out at full resolution, sometimes dramatically increasing
the readout rate while retaining the highest resolution in the region of interest (ROI). Note that
some overhead time is required to shift out and discard the unwanted pixels.
Binning
Binning is the process of adding the data from adjacent pixels together to form a single
pixel (sometimes called a super pixel), and it can be accomplished in either hardware or
software. Rectangular groups of pixels of any size may be binned together, subject to
some hardware and software limitations.
Hardware Binning
Hardware binning is performed on the CCD array before the signal is read out of the
output amplifier. For signal levels that are readout noise limited this method improves
S/N ratio linearly with the number of pixels grouped together. For signals large enough to
render the camera photon shot noise limited, the S/N ratio improvement is roughly
proportional to the square root of the number of pixels binned.
Binning also reduces readout time and the burden on computer memory, but at the
expense of resolution. Since shift register pixels typically hold only twice as much charge