In Huygens Pro there are two tools to measuring a PSF from latex beads. Which to use?
When measuring a point spread function from an average image of latex beads, there are two roads to take, one in which a sphere is generated separately and then used as a PSF in the MLE restoration of the average bead image and another, more automated method using the "reconstruct PSF" command directly from the average bead image. According to the Owners Manual, the automated method takes into account microscopic parameters as well. What exactly does the automated method take under consideration because it is quite clear that the two techniques produce differing point spread functions.
The automated method generates the sphere on the fly and determines the background automatically. The rest of the procedure is the same, except that the non-automatic method has a different procedure to compute the output size and allows setting of the iteration stop criterion.
The reason why we keep the non-automatic method around is that sometimes one needs control over the background, exact dimensions and number of iterations. The downside is that it is far more error prone, especially with respect to the background and sphere size. The automatic background determination is since version 2.1 very robust, rarely fails, and hard to improve on 'manually'.
Keywords: measuring bead psf reconstruct
Categories: Faq Deconvolution, Huygens Faq, Imported Faqs
Platforms: Irix Linux
Related products: Hu Pro
