Improving Image Resolution

[coolraiser103]

When it comes to planetary imaging, I seem to succeed in the processing part. An image of mine from July 4, 2014 shows what I mean.

(I placed it into the attachments since it doesn't want to go into this post.)

It was taken using a Vintage Celestron 5, and a Neximage 5 for the camera. It was stacked and processed in Registax 6, and finalized in Paint.Net.

The issue I have with it is the resolution. The seeing that night was wonderful, but with such a small image the details just can't come out. I'm unsure of how to proceed without spending too terribly much on this. In addition, I feel that using a barlow lens isn't often a great idea due to the increased blur and decreased light input, overall creating noise.

Perhaps I might be making the wrong conclusions? What is a decently inexpensive method to increasing image resolution in planets?

As always, be sure to ask questions for clarification!

Saturn July 3 2014 1055PM.bmp

[kalasinman]

The software for the camera offers choices of resolution. Have you tried changing the settings? You could try Firecapture. It has a feature that selects an area of interest and limits your image to that, in effect zooming in.

[MiguelMJr]

Hello coolraiser103,

When you say "increasing image resolution" i assume that you mean that you want to see more detail of small scale structures in the final image. If so, the only way that you can do this is to get a bigger telescope.....with a larger aperture.

Your telescope is capable of resolving about 1 arcsec, the full width of Saturn and it's rings are only about 37 arcsecs.

Here is some information that may help to explain... http://www.astro-imaging.com/Tutorial/Calculations.html

Miguel

[MiguelMJr]

Try imaging the moon with your setup, you should be able to get some real nice detail. Try different binning.

Miguel

[coolraiser103]

I suppose another way to think of it might be to increase the image size itself. Of course, in order to do that, I'd need to increase my focal length...

Looking at the data for Saturn in calsky for July 4, 2014, it appears that Saturn and its rings were 41.5" wide at that time. It is 116 pixels wide in this image, so that means each pixel covers about .35". Given that the Cassini division was perhaps less than one arcsecond wide at this point, it would make sense that my image shows it as being about 2 pixels wide. What I gain from this is that increasing the focal length with a barlow would simply blur the lines between contrasting areas, since my telescope has a resolving power of one arcsecond. Thus the result would be no different as if I had simply resized the image in Photoshop.

So it does seem, then, that my images might not get too much nicer until I can purchase a larger scope.

On a slightly related note, I'm still not quite sure about the function of binning. I've been into the imaging scene for a little over a year now, but binning isn't something I've dabbled into yet.

[MiguelMJr]

Binning is simple using several photosites (not pixels) to create a larger photosite. In your case if you were to use a 3x3 bin then one pixel (in the image) would cover (3*0.36) 1.08 arcsecs.

A 3x3 bin would combine the information of 9 photosites (3 wide x 3 high) into one pixel. The image would be smaller but the "resolution" would be the same.

[ronin]

Drop a 2x or 3x barlow in the path - this does seem to be normal.

There is a good image of Jupiter that used a CPC 9.25 and a 2z teleextender (The TV thing if I have the name wrong)

You do need to use a good one for a good final image but assuming you have a barlow at hand give it a try and see how you get on.

[pete_l]

Let's review.

The Celestron 5 is a 127mm aperture F/10 scope.

The Neximage 5 is a 5MPix camera with 2.2µ pixels.

This gives an imaging scale of 0.36 arc-sec per pixel.

Now, it's difficult to quantify your "wonderful" seeing, but as a benchmark, the professional obsy near me (Calar Alto) which houses the biggest 'scope in mainland europe at an altitude of 2,000 metres has a "seeing" figure of 0.8 arc-sec [ ref: http://www.caha.es/CAHA/MISC/seeing.html ]

So even if your seeing was a truly exceptional 1 arc-sec, you'd still get 3 pixels of blur in your images, simply due to atmospheric disturbance. Increasing the aperture or focal length won't change that basic limitation.

You could try two things: both are low cost, but will take extra time.

The first is to employ "lucky" imaging with short exposure times. The seeing can have moments of great clarity in amongst the usual blurry stuff: the trick is finding them. You could try stacking only the best of the best frames and comparing the results with what you normally get.

.

The other technique is drizzle processing of your images. If done carefully, this can improve the apparent resolution. It's available as an option in Registax but it can be hit'n'miss. Some people claim good results, and some reckon it doesn't help much. I would suggest that is down to the original quality of the drizzled / stacked images.

Either way, there are certain basic, physical limitations that mean you can't pull data out of an image when it isn't there in the first place. However, some experimentation could give you subjectively better looking images. Or you could move to the top of a mountain.

[Science562h]

I have a Neximage 5 and it's way better than the basic older Neximage Solar System Imager. That's a good Saturn pic bud. You did a real good job, with the NEX5.