Xbox Webcam Advice

[Hightower]

You should be fine with manual tracking for planetry work.

Sounds to me like your webcam is not in focus. The webcam will have a different focal point to what you have with an eye piece in. Stick the webcam in, and use a home made Bahtinov mask to get the correct focus.

[xx Ash xx]

Hi,

Apologies as I know from reading through this forum that this subject has been covered quite extensively but since there are so many different scopes / focal lengths / apertures / mounts etc it can get a little confusing.

I believe I have a Helios Skywatcher 150 (f=1000mm d=150mm), an EQ3-2 mount, 10mm & 20mm Eyepieces & a 2 x barlow which I had as an 18th Birthday present 12 years ago and for most of its life has sat unused in the loft..... https://www.dropbox.com/s/kdt5r3biffbpwze/DSC_3754.JPG

Since taking up photography recently, i've found myself heading towards astrophotography more and more...

I recently purchased an xbox webcam and have modified it so that I can attach it to my scope which included removing the standard IR filter.

It was my first time using that setup lastnight and I was a little disappointed by the results. I tried to record jupiter but all that I have to show for it is what can only be described as a blurry blob. The total exposure time of the video was about 25 seconds. Is this long enough to get any kind of detail?

I've installed PIPP & Registax but still I couldn't pull out anywhere near as enough detail as I would have liked.

I also tried imaging the moon, but since I don't have a motor the moon moved through the frame quite qucikly within about 30 seconds. Putting that video through registax didn't really work either, which i'm guessing is down to the moon moving through the frame so quickly.

Is a motor drive essential for this kind of work or would I be able to get away with manual tracking?

[xx Ash xx]

Home made? I honestly wouldn't know where to start but am willing to try anything.... I had a quick look here http://astrojargon.net/MaskGenerator.aspx and filled out the first three pieces of info and the produced mask looks too small..

I entered

1000mm for the focal length

150mm aperture width

I'd got the webcam hooked up to my laptop and was viewing the footage through Sharpcap and it looked as infocus as I could get it. But obviously not in focus as was requried.

[JamesF]

How many frames were you stacking? Ideally you want to be able to choose from as many as possible. For Jupiter you could probably capture frames over a period of two to three minutes for stacking.

Experimenting with wavelets in Registax may be necessary. It's hard to suggest what's right there. Spending time tweaking them is probably the only way to find what works.

It may also be easier if you start without the barlow. The more you increase the image scale the harder it gets.

James

[Hightower]

Home made? I honestly wouldn't know where to start but am willing to try anything.... I had a quick look here http://astrojargon.n...kGenerator.aspx and filled out the first three pieces of info and the produced mask looks too small..

I entered

1000mm for the focal length

150mm aperture width

I'd got the webcam hooked up to my laptop and was viewing the footage through Sharpcap and it looked as infocus as I could get it. But obviously not in focus as was requried.

Try 1000, 150, and 15 for the edge. I've just checked and it should fit inside a 150 (rest on the spiders for the secondary). Simply print it on to some card, and get a craft knife and cut all the white bits out. Metal ruler might come in handy too, to keep your lines straight.

How did you focus the telescope for the image you took? Did you focus with an eye piece and then just swap out the camera? Or did you try focusing with the camera on the laptop screen?

As James has said, longer videos will generally produce better images, but it does sound like a focusing problem if it is totally washed out - you should still see some sort of detail.

[xx Ash xx]

Try 1000, 150, and 15 for the edge. I've just checked and it should fit inside a 150 (rest on the spiders for the secondary). Simply print it on to some card, and get a craft knife and cut all the white bits out. Metal ruler might come in handy too, to keep your lines straight.

How did you focus the telescope for the image you took? Did you focus with an eye piece and then just swap out the camera? Or did you try focusing with the camera on the laptop screen?

As James has said, longer videos will generally produce better images, but it does sound like a focusing problem if it is totally washed out - you should still see some sort of detail.

I initially focused with the eyepiece first which I then replaced with the webcam and thinking about it I had to focus it again. The primary reason for me using the eyepice first was inorder to line the scope with the finder scope so that I knew what I was aiming at.

[xx Ash xx]

How many frames were you stacking? Ideally you want to be able to choose from as many as possible. For Jupiter you could probably capture frames over a period of two to three minutes for stacking.

Experimenting with wavelets in Registax may be necessary. It's hard to suggest what's right there. Spending time tweaking them is probably the only way to find what works.

It may also be easier if you start without the barlow. The more you increase the image scale the harder it gets.

James

I'd say for Jupiter, I was recording for about 25 seconds at 30 fps which would give me about 750 frames (This is from memory as i'm at work at the moment and my files are at home)

So as long as I keep Jupiter in the field of view of the webcam I can manually track it and get good results?

[Hightower]

I'd say for Jupiter, I was recording for about 25 seconds at 30 fps which would give me about 750 frames (This is from memory as i'm at work at the moment and my files are at home)

So as long as I keep Jupiter in the field of view of the webcam I can manually track it and get good results?

Yeah. With DSO imaging and DSLR camera you leave the mirror open, allowing more light in. If you move the scope while doing this you'll end up with a blurred image.

However, with a webcam and planetry you don't leave it open, it just takes a standard video instead. It doesn't matter whether Jupiter starts at the top and then drifts to the bottom as you track it, as the software you use to stack it will align and recentre it for you. You could even use 30 x 25 second clips stitched together to give you more frames and hopefully more detail.

[xx Ash xx]

Yeah. With DSO imaging and DSLR camera you leave the mirror open, allowing more light in. If you move the scope while doing this you'll end up with a blurred image.

However, with a webcam and planetry you don't leave it open, it just takes a standard video instead. It doesn't matter whether Jupiter starts at the top and then drifts to the bottom as you track it, as the software you use to stack it will align and recentre it for you. You could even use 30 x 25 second clips stitched together to give you more frames and hopefully more detail.

Yes, because a dslr in that configuration with the mirror open is only gathering light for one frame whereas a webcam is taking multiple frames per second. Is it a better idea to reduce the fps of the webcam whilst increasing its exposure?

[Hightower]

Yes, because a dslr in that configuration with the mirror open is only gathering light for one frame whereas a webcam is taking multiple frames per second. Is it a better idea to reduce the fps of the webcam whilst increasing its exposure?

Yes, to an extent. You don't want to have it over exposed, and when talking of photography using a webcam you need multiple frames. The more the better, so don't turn it too low on FPS. The software will decide (along with your input if you want) which frames are classed as poor, and which frames will help make up the final image.

I think the default on Registax (from memory) when you select which frames is based on lowest quality, and the level is set to 95% (which can be changed). Now this doesn't mean 95 out of 100 frames will be binned. It means that if a picture isn't within the 95% threshold quality wise (compared to the best frame you have) it will be binned. If you have a really good video, this means that only 70% could be binned. If you have a bad video it could mean 99% is binned.

Registax works very well out of the box, but it works even better with a minimal of tweaks - it's definitely worth playing about with it and trying to tease quality out of images. Hey, it's kind of like a sub-hobby for all those cloudy nights we have!

[JamesF]

if the image does drift about a fair bit it's worth looking at using PIPP to crop the frame around the target and centre it. That makes the stacking programs life rather easier.

James

[xx Ash xx]

Yes, to an extent. You don't want to have it over exposed, and when talking of photography using a webcam you need multiple frames. The more the better, so don't turn it too low on FPS. The software will decide (along with your input if you want) which frames are classed as poor, and which frames will help make up the final image.

I think the default on Registax (from memory) when you select which frames is based on lowest quality, and the level is set to 95% (which can be changed). Now this doesn't mean 95 out of 100 frames will be binned. It means that if a picture isn't within the 95% threshold quality wise (compared to the best frame you have) it will be binned. If you have a really good video, this means that only 70% could be binned. If you have a bad video it could mean 99% is binned.

Registax works very well out of the box, but it works even better with a minimal of tweaks - it's definitely worth playing about with it and trying to tease quality out of images. Hey, it's kind of like a sub-hobby for all those cloudy nights we have!

Thanks...

So, its about finding a happy medium, too many fps and over exposed but not too low as not enough light would be coming in....

Switching back to my comments about capturing the moon, is is possible to obtain good results with manual tracking is a motor the best way to go? When thre webcam is attached to the telescope, I can only see about a 10% of the moon so I don't think manually tracking will work for me especially since the moon moves through the frame quite quickly; My hands aren't that steady.

Is there a way to zoom an Xbox Webcam out?

[xx Ash xx]

This, rather emabarrasslingly is what I managed to capture of Jupiter last night https://www.dropbox.com/s/4nuht5nisukhqed/Capture%2019_2_2013%2019_47_46.avi?m

and this is the resulting image pulled out using PIPP & Registax https://www.dropbox.com/s/rg94bzhzf45t7iq/jupiter.bmp?m

[JamesF]

I think you need plenty more gain than that, possibly combined with longer exposures.

On the transforms menu in SharpCap there's a "histogram" entry. If you select that you can see the histogram for the capture. I'd start by recording at 10fps, maximise the exposure time to allow that (I think it turns red if the exposure time won't allow 10fps) and then push the gain up until the histogram has data about three quarters of the way across.

James

[nephilim]

Hi, most people make a mosaic with the moon ( you can use a focal reducer to reduce the moons size but these are quite expensive) , basically taking a series of avi's of all the parts visible , stacking to produce an image & stitching the stacked images in a program such as imerge, manual tracking is difficult as your constantly adjusting settings, focus and THEN manually tracking it, it is possible but much easier with motors, u dont 'need' the dual axis type as the basic ra drive will suffice.

Steve

[Hightower]

So, its about finding a happy medium, too many fps and over exposed but not too low as not enough light would be coming in....

It's not really about the light coming in when talking of FPS, that's your exposure setting. For FPS, you need to find a medium between capturing as many frames as possible in as short a time as possible, but also not having the frame compressed too much. The webcam will compress frames if it can't send them fast enough, resulting in reduced quality as your capture software. I've found that around 10FPS is pretty good. A 30 second video will give me 300 frames, but they don't get compressed too much at the camera. Have an experiment - take a film as 30, and one at 10, and one at 5 and compare the differences. Pick the one that seems the best compromise, but remember that these settings aren't a "I've found a good one, that's it forever" - keep experimenting when you have time, you might just find a setting that gets a better image.

Switching back to my comments about capturing the moon, is is possible to obtain good results with manual tracking is a motor the best way to go? When thre webcam is attached to the telescope, I can only see about a 10% of the moon so I don't think manually tracking will work for me especially since the moon moves through the frame quite quickly; My hands aren't that steady.

Is there a way to zoom an Xbox Webcam out?

Are you barlowed? I would remove the barlow if so. But again with the planets, you don't need to track it perfectly.

Have a look at the image I've drawn up. This is on a simplified scale obviously.

Say you have 2 frames in your video. Frame 1 shows the left hand side of the moon. Frame 2 wasn't tracked quite right so shows a slightly different part of the moon. You then stick them in the stacking software, which aligns and rotates frames, and finds these two frames overlap, and they both add something to the final image. So your final image would come out, cutting out any bits that didn't fit.

So, yes, you need to track a little bit, but it by no means needs to be perfect as the software does so much to help you.

[Hightower]

Yeah, after watching your video - it's as James says. Take a look at your gain settings and exposure settings. There's no where near enough light getting in. But looking at your image, I can just see some detail, so that shows you how good a job Registax does, even with such a dull video.

[xx Ash xx]

Hi, most people make a mosaic with the moon ( you can use a focal reducer to reduce the moons size but these are quite expensive) , basically taking a series of avi's of all the parts visible , stacking to produce an image & stitching the stacked images in a program such as imerge, manual tracking is difficult as your constantly adjusting settings, focus and THEN manually tracking it, it is possible but much easier with motors, u dont 'need' the dual axis type as the basic ra drive will suffice.

Steve

I understand the theory behind polar aligning my mount and then using the RA gears inorder to track a DSO but what I'm not entirely sure about is the ability to track the moon since its path isn't set in stone is it as opposed to stars and nebula etc

[nephilim]

I know what u mean, when I used my old ra motor on my eq2 I just had the speed settings as normal (ie select north & switch on) and it usually went well, but that does depend on the accuracy of ur PA, I only used a rough PA ie pointed the leading mount leg to Polaris so I had to tweak very slightly in dec now & again but in ra it was fine.

Steve

[xx Ash xx]

Hi, most people make a mosaic with the moon ( you can use a focal reducer to reduce the moons size but these are quite expensive) , basically taking a series of avi's of all the parts visible , stacking to produce an image & stitching the stacked images in a program such as imerge, manual tracking is difficult as your constantly adjusting settings, focus and THEN manually tracking it, it is possible but much easier with motors, u dont 'need' the dual axis type as the basic ra drive will suffice.

Steve

I was looking at a cheap eq2 motor drive like this one http://astrobeano.blogspot.co.uk/2012/01/simple-motor-for-skywatcher-eq2-mount.html but wasn't sure whether i'd be able to attach it to my eq3-2 mount....

[jakeybob]

I'm new at this too, and seem to be encountering similar problems. (I'm using a 200P Dobsonian on push-to alt/az mount, but the same Xbox webcam as you).

Camera

I had a bit of trouble achieving focus with the webcam at first. I found that I had to put it right in the focus tube, to the extent that I loosened the thumbscrews to be only half holding the webcam's tube in order to get some play in the focussing. However, it should be easy to tell by eye by focussing on the moon. This should be obviously in or out of focus, even on a dimmed netbook screen. From the look of the video though I'd agree with the others who said it's more about tinkering with the gain and exposure. I found my webcam likes to drop frames all over the place when the exposure setting is too low, so I tend to use that and the Histogram filter in Sharpcap to tune it. Having said that, I think I maybe tend to overexpose ...

Software

My setup involves leaving the scope stationary and letting the moon/planet drift through the field of view. See here for a movie I took of Jupiter the other night.

http://www.flickr.com/photos/jakeybob/8493422258/

I use a piece of software called Castrator to centre/crop the avi file so that it looks like this:

http://www.flickr.com/photos/jakeybob/8493420362/

This seems to make it a lot easier for the stacking software to deal with, and has the added benefit of massively reducing the file size. In the end I got this picture out of that capture:

http://www.flickr.com/photos/jakeybob/8493398908/

As far as lunar shots go my tactic so far has been to let it drift through the frame (http://www.flickr.com/photos/jakeybob/8492671492/) and then use VirtualDub to crop the avi so that only a few hundred frames remain around the specific area I want to image - i.e. so the bit I want to make a picture of is always in the frame somewhere. Then I define that area as a "region of interest" in Registax, which seems to force registax into behaving itself So for example I used the last few hundred frames of that moon movie to stack into this picture: http://www.flickr.com/photos/jakeybob/8492724302/

Sorry if I'm prattling on - I'm finding it useful writing all this down

[nephilim]

I was looking at a cheap eq2 motor drive like this one http://astrobeano.bl...-eq2-mount.html but wasn't sure whether i'd be able to attach it to my eq3-2 mount....

I'm not sure that would fit the EQ3/2 but i may be wrong. The best thing to do would be to ring FLO (the sponsers at the top of the home page here & ask them, they'll be more than happy to advise you.

Steve

[nephilim]

I'm new at this too, and seem to be encountering similar problems. (I'm using a 200P Dobsonian on push-to alt/az mount, but the same Xbox webcam as you).

Camera

I had a bit of trouble achieving focus with the webcam at first. I found that I had to put it right in the focus tube, to the extent that I loosened the thumbscrews to be only half holding the webcam's tube in order to get some play in the focussing. However, it should be easy to tell by eye by focussing on the moon. This should be obviously in or out of focus, even on a dimmed netbook screen. From the look of the video though I'd agree with the others who said it's more about tinkering with the gain and exposure. I found my webcam likes to drop frames all over the place when the exposure setting is too low, so I tend to use that and the Histogram filter in Sharpcap to tune it. Having said that, I think I maybe tend to overexpose ...

Software

My setup involves leaving the scope stationary and letting the moon/planet drift through the field of view. See here for a movie I took of Jupiter the other night.

http://www.flickr.co...bob/8493422258/

I use a piece of software called Castrator to centre/crop the avi file so that it looks like this:

http://www.flickr.co...bob/8493420362/

This seems to make it a lot easier for the stacking software to deal with, and has the added benefit of massively reducing the file size. In the end I got this picture out of that capture:

http://www.flickr.co...bob/8493398908/

As far as lunar shots go my tactic so far has been to let it drift through the frame (http://www.flickr.co...bob/8492671492/) and then use VirtualDub to crop the avi so that only a few hundred frames remain around the specific area I want to image - i.e. so the bit I want to make a picture of is always in the frame somewhere. Then I define that area as a "region of interest" in Registax, which seems to force registax into behaving itself So for example I used the last few hundred frames of that moon movie to stack into this picture: http://www.flickr.co...bob/8492724302/

Sorry if I'm prattling on - I'm finding it useful writing all this down

Thats a really good sharp image of the moon mate, with Jupiter I think you may have slightly 'overdone' the wavelets but still a good image.

Steve

[JamesF]

I'm not sure that would fit the EQ3/2 but i may be wrong. The best thing to do would be to ring FLO (the sponsers at the top of the home page here & ask them, they'll be more than happy to advise you.

I'm pretty sure they're not compatible.

James

[nephilim]

I'm pretty sure they're not compatible.

James

I thought not.