Author: Sebastian Rhode (Sebastian.Rhode at gmx.de) History: 2008/03/27: First version
Source: Calc_CCD_SNR.java Installation: Download Calc_CCD_SNR.class to the plugins folder and restart ImageJ. Description: This plugin allows you to calculate the SNR for two CCD cameras to compare them. Required input parameters are:
- Quantum Efficiency - quantum yield of CCD chip
- EM-Gain - multiplication factor for On-Chip Electron amplification (can be set to 1 for normal CCDs)
- Noise Factor - noise component (also called Multiplicative Noise) related to the amplification process. The factor represents the statistical variation in the overall number of electrons generated by the initial detected charge. For a stochastic gain process in a EM-CCD the noise factor = 1.41 and for normal CCDs = 1.0 and for modern ICCDs = 1.6.
- Dark Current - charge generated by thermal generated electrons. The actual dark Noise is the statistical variation of the Dark Current (sqrt(Dark Current). Dark Current can be subtracted from an image while the dark noise remains.
- Spurious Noise - Clock Induced Charge (CIC) which is independent of the exposure time. During the transfer of electrons there is a small probability of producing extra electrons through impact ionization. Is normally hidden in the ReadOut Noise for normal CCDs and ICCDs. For EM-CCDs those charges will be amplified and become detectable.
- ReadOut Noise - noise related to the actual ReadOut process. Faster readout speeds will lead to a higher ReadOut Noise. That noise can be effectively reduced to 1 by using an EM_CCD.
Other terms to know:
The plugin will calculate the graphs for the two CCDs and will display the correction factor required for a correct comparison of the two SNR curves. It will help you to make the right decision when purchasing a new CCD for you imaging system.
- Shot noise - represents the uncertainty in the number of incoming photons. It is an inherent noise (can not be overcome) and is defined by POISSON statistics.
- Correction Factor - Assume the photon flux is 100 photons per pixel per frame for an 16m CCD chip. To compare that to a 8m chip keep the following in your mind:
- 16m^2 / 8m^2 = 4 --> to compare the SNR values you have to use the x=100 photons for the 16m CCD chip and the x =100 photons / 4 (!!!) = 25.
- In other words, 100 photons/frame/pixel for a 16m chip corresponds to 25 photons/frame/pixel for an 8m chip!
- SNR= SIGNAL / sqrt( NF2 * (SIGNAL + DARK + SCIC2) + (SR2/GAIN2))
|Plugins | Home |