Cooled CCD cameras

    CDD stands for Charged Coupled Device.  A CCD is a silicon chip with wells arranged in a lattice or matrix.  Each well collects photons as electrons.  

    A computer image is a 2 dimensional matrix of pixels (or voxels).  Each pixel has a discrete value, for instance 0 to 255, which corresponds with black to white.  Typically, 0 is black and the highest possible number is white.   Each pixel is addessed by its location X and Y.  Each X, Y location corresponds to a position on the physical CCD.

    Each well location in the CCD chip corresponds to a spatial location in the orginal image.  Each well contains a number of electrons which are a direct count of the photons (brightness) at the location.  The electrons are counted and the number is assigned to the corresponding pixel.  Also, wells may be added together to increase intensity and decrease noise.  This is called "binning".

    An excellent resource for technical information on CCD cameras is the library at the RoperScientific web site.

    Film and cooled CCD cameras have approximately the same sensitivity.   Sensititively means being able to detect a weak signal.  If there is a signal to detect, it can be detected by lengthening the exposure time.  More time allows for more photons to be collected.  For these weak signals, film and CCD chips require similar long exposures.  However, film must be developed whereas the post-processing time limitation for digital imaging is the speed of the computer.

    The CCD chip has a much larger dynamic range than film which can be especially useful for biologists who need to directly compare low abundance and high abundance in the same image.  A digital imaging system's dynamic range is 2 1/2 to 4 (or more) orders of magnitude depending on the instrument.  Film's dynamic range is 1 1/2 to 2 orders of magnitude.  In practical terms, this means that with digital imaging it is possible to obtain more intensity information before being limited by saturation (or clipping of weak signal).  Instead of taking two or three (or more) film exposures to capture a wide dynamic range, only one digital image is needed.

    In the figure below, the gray bar represents more than 4000 discrete intensity values.  One digital exposure is sufficient to encompass the entire dynamic range.  Three film exposures are required to record the same range, and each exposure can capture only a narrow portion of the range.

CCD:

1

1.  One exposure can contain information of very weak to very strong signal without clipping or saturating respectively.

Film:

1.
2.
3.

1.  Exposure for the weak signals which cause the mid-tones and brights to be fully saturated.
2.  Exposure for the mid-tones which does not detect weak signal and fully saturates brights.
3.  Exposure for the brights which does not detect ("clips") the weak and mid-tones.

    Digital imaging is linear.  Film has a logarithmic response.   Therefore, digital is superior for intensity quantification.  From very weak signal to very strong signal, there is a 1:1 correspondence between intensity, or number of photons, and number stored in the computer.  Film's response, on the other hand, is on a curve and tends to have low contrast at one or the other extreme of its dynamic range limitation.

   Next:  What is binning?

Any dead links or other comments?  Please contact Michael.