Page 9, a bit of details on quantum effeciency.

sections/s1Basics.tex

@@ -201,7 +201,7 @@ \subsection{Image Bit-Depth}
 
 \subsubsection{Guide Line for the Choice of Bit-Depth}
 
-In fluorescence microscopy, the choice of whether to use higher bit-depth format depends on a balance among the intensity of excitation light,  the signal intensity (quantum efficiency), the sensitivity of camera sensitivity and the gain. If the signal is bright enough then there would good S/N that you could simply use a lower bit depth but if the signal is too bright then you might need to use a higher bit depth just to avoid saturating the bits. This balancing could also be controlled by changing the gain of the camera. In the end, the choice of bit depth depends on what type of measurement you want to achieve. 
+In fluorescence microscopy, the choice of whether to use higher bit-depth format depends on a balance among the intensity of excitation light,  the emision signal intensity, the sensitivity of photon detector, quantum efficiency\footnote{Quantum effeciency is a measure of proportion of photons converted to electric impluse. For example, QE of photographic film is ca. 10\%, that of human eye is 20\%. Recent CCD allows 80\%. } and the gain. If the signal is bright enough then there would good S/N that you could simply use a lower bit depth but if the signal is too bright then you might need to use a higher bit depth just to avoid saturating the bits. This balancing could also be controlled by changing the gain of the camera. In the end, the choice of bit depth depends on what type of measurement you want to achieve. 
 If you only need to determine the shape of the target object (``segmentation''), you might not need a higher bit depth image as you could try to adjust the imaging condition rather than increasing the file size of the image. 
 On the other hand, if you are trying to measure protein density, a higher bit depth allows you to have more precise measurements so a larger bit-depth is recommended. A draw back is that it takes longer time for data transferring as the bit-depth becomes larger. This may in turn limits the time resolution of image sequences.
 The balancing between imaging conditions and the type of  analysis afterward is very much coupled and should be thought well before your experiment. More details on this discussion could be found in microscopy textbook such as \cite{Pawley2006}.