User talk:Default007

PSF and Nyquist rate

As you can see, I thought you fluorescence microscopy links were opaque, obscure, and off topic, so I removed them. If I, as an expert in these topics, can't make sense of the linked material, what good is it going to do a regular reader? Perhaps it belongs on a page about fluorescence microscopy, where the needed concepts are defined. Dicklyon 20:54, 17 June 2006 (UTC)

You're probably right! That link only without further explanations would be very obscure. But also think that plenty of biologists use microscopes today and, without being experts in image processing, they need to be aware of what the Nyquist rate is. This is just an example, but the Nyquist rate spreads more and more to many different fields, from which wikipedia readers may come to this article. I think that wikipedia articles should be able to include references to many different fields (that's why hyperlinking is so useful). What do you thing is best? To include a new section inside this article, or to link to a new article, named "Nyquist rate in microscopy"? Default007 16:10, 18 June 2006 (UTC)

A section on Nyquist rate in diffraction-limited optical imaging would make sense, including a few words and a link about microscopy in particular. Dicklyon 16:29, 18 June 2006 (UTC)

OK, thanks, I'll think about that! Another point is: despite I agree with you about the link on the Nyquist rate article as it is currently, I disagree about its inconvenience in the PSF article. This other article is not much evolved right now, and most of its current contents refer directly to microscopy. That's why I think that a [PSF calculator for microscopy] makes much sense here. In any case I don't think that concepts like Numerical aperture, Wavelength or Refractive index, on which the calculator is based, are so complicated or specific to microscopy... (but they are also explained there!). So if I convinced you, I would include the link again here...

But that calculator page doesn't do anything to teach you what you don't know. The first choice of microscope types will be complete nonsense to people not in the field, so the rest is not very useful. If it had some diagrams and formulas to help you choose and say what it's calculating, it would be useful. But it's just opaque as it is, and therefore not a page that anyone can learn from. At least, that was my impression at quick look. Dicklyon 18:02, 18 June 2006 (UTC)

OK, I see, you just gave a quick look. But instructions actually followed right below the calculator and all items are linked to descriptive articles that further explain what are they about. It is not indeed as a completely self descriptive article, it is just a calculator inside another wiki that gives plenty of explanations. I already edited that wiki article and put the instructions at the very begining. I think that calculator is intended for 1) expert microscopists, to calculate the ideal sampling rate and 2) beginners in the field who still have to read a lot, but to which seeing how a PSF looks like would be very educative. Moreover general wikipedia readers of the PSF article could find it useful just to see what a PSF is. That's what I think that an external link from the wikipedia PSF article is very pertinent. I would appreciate that you take another look at the calculator and reconsider the deletion of my edition. Whenever you find it OK, I would put it back. Default007 19:33, 18 June 2006 (UTC)

That's an improvement. Go ahead and link it (you don't need my approval, but thanks for considering my opinion). It's still a bit opaque for my taste. I can't see what formula it's using, or what criterion for establishing a bandwidth limit. It would be good to say more precisely what it's doing. And say how big (in nm) th pixels are in the PSF plot. Also, could you go ahead and also calculate depth of field while you're at it, using some conventional criterion like almost as sharp as the best diffraction-limited focus sharpness? Dicklyon 19:48, 18 June 2006 (UTC)

The PSF image size is given in micrometers. It is calculated at Nyquist rate, that is given in nm. The image is upscaled (3x) after calculation to allow a better view of the sampling density. Therefore a pixel in the final image is not an original pixel, but the anisotropic pixelation is clearly seen.