Quote:
Originally Posted by spearo
Nice image Jase as usual.
question: using different instruments, wouldn't that give subs with the target at slightly different scale (for example, exagerating the point something shot with my c9.25 and the c14 will vary because of the different focal lenghts.)
Did you have to rescale he images to stack them?
if so how do you do that (i've often wondered how to do it to achieve a perfect match) ?
frank
|
Thanks Frank!
Yes, you're right, different instruments produce different image scales. Though we are not talking about a huge difference. Matching data say 500mm with 3000mm becomes a challenge. It can be done, but takes some work. To answer your question, its software that works its wonders. Registar is probably the easiest to use and pretty much designed to align subs as best a possible, but I'm progressively not using this considering the the software doesn't output 32bit floating format. Will read it, but doesn't write it. Not a problem if you're working in the 16bit space already, but I wouldn't trust Registar to scale an image from 32bit to 16bit. I have no idea what type of scaling it is doing. Many software packages will do the scaling. In this instance I used CCDStack and resized the images so they are where close, then used the CCDIS/P plugin. The initial resize is generally not needed, but I do it anyway. Consequently I note that a new version of the plugin for CCDStack was released yesterday. 2.15 if I recall correctly. In short the rescaling of images is no different from when you need to upscale 2x2 (binned) RGB data to match that of 1x1 luminance. You simply need a software package such a Registar, CCDStack, MaximDL etc. (there are heaps) to do a star matching algorithm to scale the images. Hope this explains a few things. Tip, scaling works exceptionally well if you are down scaling a luminance...say to match a wide field image. I would only recommend upscaling RGB. Luminance upscaling typically results in a loss of resolution. This doesn't matter for RGB as its not the resolution you're after but the chrominance information to match against the high resolution luminance.
Quote:
Originally Posted by atalas
Nice work Jase!
|
Cheers Louie