Astroimage processing

Re: Saturn: webcam image processed with Registax

TeleNoob — Wed, 23 Apr 2008 20:33:16 GMT

I took a look at your image as I have been working on the same subject, Saturn and processing with Registax. I'm a newb, so asking some of the imaging experts some advice here is so helpful. I noticed your image has a moon in it, probably Titan based on its orbital distance, but I'm not sure. This is of some interest because unlike Saturn it is not over-exposed.

What I do when trying to find an object is crank the exposure time and gain up, to help identify when its in the field of view, but then turn these down until the image becomes clear, not saturated. This has to be done with Automatic Gain Control on manual. If its in Auto, the camera sees mostly black so it cranks the gain up to maximum every time. Planets are actually pretty bright so they require the gain and exposure to be lower, and this usually needs to be done with AGC in manual.

But getting back to the moon, I zoomed and processed your shot and noticed some strange artifacts around it, some kind of noise or nebulous cloud. There seems to be an arc or filament on the left side of the moon image. I doubt that this is real, but since the image was apparently procesed in Registax it must be there in the majority of frames. So what is it?

Here in this pic, it first shows the moon un-processed, then with some contrast and filtering, then with the colours inverted.

Re: Saturn: webcam image processed with Registax

darticus — Mon, 31 Mar 2008 17:54:00 GMT

Really searching.

Last night my wife came out and opened the steel door next to me, the scope, and computer. The image on the computer screen faded away to gone. It than came back slowly when the door closed. The door opens in away from the equipment. We did this several times and each time we lost the image. Could this cause the noise I have in my images when the door is closed? Anyone see disturbance from metal? Maybe I have to move the scope.

How about settings seems like I need gain high. Maybe its the door?

Re: Saturn: webcam image processed with Registax

chipdatajeffB — Mon, 31 Mar 2008 15:30:04 GMT

You can do it simultaneously with R4. With R2 or R3 you must do them separately then combine the resulting single frames (a pain).

Which version are you using?

In looking at your image, it's hard to be sure but I think the barlow has caused you too much magnification for the seeing. I could tell better from the AVI file. How big is it? If it's small enough, put it in the Yahoo Group Files area and send me a link to it so I can check it out.

Or send it to me as an email attachment and I'll process it and tell you what steps I used.

Re: Saturn: webcam image processed with Registax

darticus — Mon, 31 Mar 2008 15:23:25 GMT

Are you trying it 3 different ways. Is there a way to mark it with 3 boxes at one time. See my diaster pic. http://tech.ph.groups.yahoo.com/group/neximage/photos/view/b603?b=6

Re: Saturn: webcam image processed with Registax

chipdatajeffB — Mon, 31 Mar 2008 15:00:38 GMT

It may help to try it first without the barlow, as focus and exposure might be easier. Once you have that mastered, try it with the barlow. It will be significantly more difficult to focus using the barlow (that's just the nature of focal length).

The image below shows where I normally locate alignment boxes in Registax:

When the planet is aligned well to show the shadow of the rings passing across the body of the planet, that's a good place to center an alignment box if you can focus well enough to show that shadow. If not, then either use two alignment boxes positioned in the gaps between the rings and the planet, or one single alignment box large enough to surround planet and rings and centered on the planet.

Note the outer boxes in the image above are centered on the Cassini Division. With the rings so nearly closed right now you can't do that -- center them in the gap between the rings and the planet instead.

Which version of Registax are you using? I find that Registax 4 has confused the Limit option that was clearer in R3. If you're only seeing 5% good frames, then one or more of the following Registax settings needs checking:

Your Quality Estimator alignment option may be set to Gradient instead of Local Contrast. When aligning images of Saturn, use Local Contrast.
You may have your alignment box/boxes set in a non-optimal place.
You may have the boxes set too small. Note the boxes in my image above may be too small for current views of Saturn. Use larger boxes.
After Alignment, you should see an indication of how many frames it has selected as "best". You can change this before you hit Limit (and then reduce it later if you like, when you get to Stack). To change the number of frames, move the vertical bar in the Alignment graph to the right by clicking and dragging the small pointer in the horizontal line at the bottom of the Registax window to the right. You should see the number of frames to be processed increase as you move the pointer to the right. I move it to include about half the frames, as a default. If I see the graph gets very "ragged" toward the right, then I stop adding frames when the Limit line on the graph (vertical line) approaches the beginning of the raggedness.

After you click limit, use Optimize and NOT Optimize and Stack. Optimize and then pause when it's done. Click Stack, and then look above the window to the options bar. Click Use Stackgraphs.

When you use multiple alignment points, only one stackgraph is displayed at a time. Click the numbered box at the lower right-hand corner of the stackgraph window to change between alignment points. Remember that some alignment points may align and optimize differently, so you may want to use more or fewer frames for some alignment points than for others.

To change the number of frames used with a given alignment point, display the stackgraph for that point and then move the sliders on the stackgraph window to move the lines to appropriately exclude the frames you don't want. A good rule of thumb is the 80/20 rule: cut off 20 percent of the frames using the quality or difference sliders. In good seeing, you may not want to cut off many using the difference slider, since there won't be as much difference frame to frame.

The Quality slider is the horizontal slider below the Stackgraph box. The Difference slider is the vertical slider to the left of the Stackgraph box.

Experiment with these settings when the seeing is good, as it will make a more noticeable difference in the final stack when there is more detail to work with.

I've edited your post, above, to include an image. Click Edit on that post to see how it works. You can substitute a URL for one of your images to display it. To get the [ img ] tags to appear, you can either type them or click on the "tree" icon above the Post window.

Re: Saturn: webcam image processed with Registax

darticus — Mon, 31 Mar 2008 14:37:26 GMT

Keep trying but can't seem to get a good image of saturn. Using a C8 with a 2x barlow and Neximage. Had about 2000 frames but it seemed to only stack 5% of the total to get this. Maybe I should be picking a different section other the the center over the bands to set up to.

http://tech.ph.groups.yahoo.com/group/neximage/photos/view/b603?b=6

Darticus: I edited your post to include the image below:

so you can see how this works. Just click Edit (below) to display your post. Note the [ img ] and [ /img ] tags, with the image URL between them. To get those tags into a post, either type the directly, or click on the tree icon above the Post window. The tree icon is the one I used for this sample image. Hope that's clear!

Re: Saturn: webcam image processed with Registax

tkerr — Mon, 25 Feb 2008 23:55:10 GMT

Ok Jeff, I've got clear skies tonight, So I'm going to try your method through my 6" and see what I come out with.. Too bad the tilt is more edge on this year..

Have A Nice _________

Re: Saturn: webcam image processed with Registax

maryccc — Fri, 22 Feb 2008 16:55:28 GMT

OMG those are all so great. I only dream that I can get one like that of saturn. I have a phillips webcam and have tried but can't find saturn in it. The moon is just too big in it. I may try again.

Re: Saturn: webcam image processed with Registax

chipdatajeffB — Mon, 18 Feb 2008 05:15:34 GMT

Those lines in your MAP image of the Moon are the result of how Registax combined the registered images based on your choice of alignment points.

I have this happen when:

I pick too few alignment points (except for a single one!)
I have them too close together
I have them too far apart
I have the "feathering" option set too coarsely

I find it hard to predict which of these will cause me a problem, so I normally notice it only when it's too late.

For planetary imaging, I find that most often the fix is to make more alignment points and space them roughly equally around the areas of detail. But this depends a lot on how much detail there is in the image.

For lunar imaging, I find the best method for doing multiple alignments is either:

Do the AVI multiple times, each time selecting a single alignment point and saving the resulting stacked (but not finished) AVI, then running all the AVIs thus made into Registax at one time (reloading them) and registering them together with a single large alignment box centered on the area of the image with the best detail in it. I usually have to crop the result, because the edges will be slightly rotated or offset by Registax.
Select multiple alignment points but only in the areas where the detail is best. Again, you may have to crop the result.

I also find that picking the best frame as the Reference frame (before Limit) is the best way to ensure a good result. For that reason, I almost always now use the feature called Create New Reference to make a Reference frame from multiple good frames. I always use this method on planets, and less often on the Moon, since I first discovered it.

Hope these tips help!

There is a good Yahoo Group for Registax users now. Some of the posters there have developed their own processing tutorials, and many of those are excellent. Join that group and do a Search for "tutorial" and you'll find some recent and relevant links.

Re: Saturn: webcam image processed with Registax

theta_orionis — Mon, 18 Feb 2008 04:59:58 GMT

chipdatajeffB:

I start with autoexposure selected, to get an initial focus.

I then switch to manual exposure.

Gain is set as low as possible and still see and image good enough to focus.

Brightness is set about mid-range.

I do not make contrast or gamma corrections.

Saturation is set low enough to avoid seeing any color bits in the background.

Hue is set mid-range.

White balance for Outdoor.

I set frames per second to its lowest setting (in my software this is 15fps).

I set the exposure as low as it will go (longest exposure).

I then spend a good 10 to 15 minutes balancing the exposure against the brightness setting while maintaining a good focus.

Jeff,

I tried your suggestion and was able to get a decent looking video of the moon (by the standards for for a greenhorn :-))

However, I tried to stack in Registax using 3 points and got a funny looking image. It looks like different pieces are clipped with a scissor and sewn together. Not what I really expected to see. Here is the image:

Would you see something wrong going on with my approach? Or is it the software?

Gurus, please help.

Thanks and Regards

Re: Saturn: webcam image processed with Registax

chipdatajeffB — Sat, 16 Feb 2008 11:35:41 GMT

Rankinstudio:

I also have a few questions. I'm using a quickcam pro 5000 which has a vga CCD imager. I was wondering if anyone else has tried to shoot the planets with this cam?

I have used the Color Quickcam and the QuickCam 4000. They are not as sensitive as the Philips SPC900NC seems to be, but you should be able to get good results nonetheless.

...I'm thinking maybe im not getting it in focus right, ...

Focus is the most important thing. Without good focus, everything else can be perfect and you'll still have a bad image. I usually start my imaging sessions by focusing on a bright star. I tweak the gain and brightness down as low as possible and then focus the star to the smallest possible point. This will also tell you a lot about the seeing conditions.

Then I slew over to my target and look at focus more critically. I often refocus between AVIs, depending no the seeing and temperature. A temperature that is changing rapidly during the session will change your focus during the session. My focuser and my scope's objective cell are both temperature-compensating designs. Even so, I can see the focus change during a session which has wild swings in ambient-air temps.

Truth be told, I spend as much time focusing and refocusing as I do imaging during an average session.

... or I need more than a 2x barlow for my scope (1200mm Focal length) 8 inch diameter.

I find that my best planetary images are made at focal lengths of around 3,000mm to 4,800mm. I often go as high as 6,000mm when the seeing allows. I use mostly a 1,200mm 6-inch refractor. I prefer Televue Powermates. You really can see the difference in sharpness and contrast (image clarity) compared to lower-priced barlows.

...

Also, about the virtural dub reprocessing. Could you be more specific on that. I have the prog, but cant find the option to increase the size by a percentage.

I don't have Vdub on the computer I'm using right now. There is a menu option that allows you to select what are called Filters. When you select that option, a Dialog box appears with an option to Add filters to a list. When you click Add, you see a list of filters and one of them is a Resampling filter. Resampling changes the dimensions of an image by adding or subtracting pixels in the image according to different interpolation algorithms. A good one to use is the Lanczos filter. After you select that filter, you can input the increase or decrease in size that you want. Once you have the filter list complete (you can add any number of filters to the list, including noise reduction, color correction, resampling, etc.), you Run the filters on your image (e.g., your AVI file) or some selected portion (e.g., frameset) of it and VDub prompts you to save a copy of the new file. You then process that new (e.g., resized) file with Registax.

Re: Saturn: webcam image processed with Registax

chipdatajeffB — Sat, 16 Feb 2008 11:23:22 GMT

TheHarvester:

Final stack of about 2,000 frame movie.stacked in Registax and edited in Photoshop.

Not near as good as what Jeff does but in my opinion it is a nice start.

Actually, I think that's an extremely nice image! You have great detail buried in the data there and the stacking brought it out well. Your focus is excellent, which is hard to do with Saturn. With the rings so nearly closed right now, it is difficult to pull out the Cassini Division -- especially when the seeing is less than perfect. Good job!

Re: Saturn: webcam image processed with Registax

TheHarvester — Sat, 16 Feb 2008 05:31:36 GMT

I do not know what the setting were when I made my capture of Saturn last week but here are a couple images so you can get an idea of what to start with . This was my first attempt with a web cam (phillips Toucam) in my 8" LX200 F-10 and a 2 1/2X Powermate.

Single frame captured from a video editing program I have.

Final stack of about 2,000 frame movie.stacked in Registax and edited in Photoshop.

Not near as good as what Jeff does but in my opinion it is a nice start. At least I got the Cassini division. I had the camera set at 320x240 and next time will try to remember to set it to 640x480 for a large image. The very firs set of images I got looked just like yours.then I played with the settings to make it look like what I got in the first image.On the plus side of your image it looks like you got one of the moons though.

Re: Saturn: webcam image processed with Registax

Rankinstudio — Thu, 14 Feb 2008 18:29:49 GMT

I also have a few questions. I'm using a quickcam pro 5000 which has a vga CCD imager. I was wondering if anyone else has tried to shoot the planets with this cam? I have tried a few times and am not happy with the results. I'm thinking maybe im not getting it in focus right, or I need more than a 2x barlow for my scope (1200mm Focal length) 8 inch diameter.

Here is an example of what im able to do. (image has been cropped)

Also, about the virtural dub reprocessing. Could you be more specific on that. I have the prog, but cant find the option to increase the size by a percentage.

Thank you

Cheers!

Re: Saturn: webcam image processed with Registax

chipdatajeffB — Tue, 12 Feb 2008 18:52:48 GMT

theta_orionis:

1. The vLounge software that I received with my webcam records only .wmv format. In order to have the .avi output, I am using the QCFocus to produce the avi's to feed the Registax. I am not very familiar with the QCFocus interface but I think I did see the control settings button to maniulate the gain, exposure etc..

You may have a different driver installed. I seem to recall when I installed VLounge I had a choice of drivers. At any rate, my copy produces AVI files. I have also used K3CCD Tools to very good effect. I have not used QCFocus.

2. I do not have a Go-To mount,and the image of the planet seems to quickly drift at that high a magnification. So, by the time I can make a judgment about the right focus, it just drifts away from the FOV. To make it worse, the time between my tuning the focuser and its effect showing up on my laptop monitor is also definitely delayed.

Use a higher frame-rate setting while focusing. If you're getting jumps in the preview while focusing, it's probably because the frame rate is low. The frame rate won't have any effect on the actual focus, but a higher rate will update the screen more frequently.

By GoTo I assume you mean here a tracking mount. If you are using a dob mount, then you'll need to use the drift method (just let the target drift through the field of view and bump it back to the other side one it reaches the edge). If you set your original alignment box in Registax properly, it will exclude the frames that are too near the edge of the field of view. You can save the "Limited" file as an AVI and then reprocess it using multi-alignment points later, if you choose.