Trying to Get Decent Image of Jupiter with Webcam

[JimD]

Hi all, I need help!

I've acquired an X-Box webcam, and attached it to my E150P. I got a couple of OK shots of the moon with it, but not having much luck with Jupiter.

I don't have a motor drive yet, and I was hoping to use RegiStax to align the frames of the moving image, but I'm not sure if the basic raw AVI is not good enough or if I'm not using RegiStax properly. I've posted the best still frames I can find from a few sequences, hopefully someone can tell me where I'm going wrong. I'm using Sharpcap to capture the AVIs (setting are below each pic).

I tried using lower resolutions, but the result was similar quality, just more pixelated. Higher gains/exposure just resulted in the image being washed out.

Capture 15_02_2013 18_08_14 RAW STILL.bmp

Frame Divisor=1

Resolution=1280x960

Frame Rate (fps)=7.50

Colour Space / Compression=YUY2

Exposure=-10

Brightness=128

Contrast=32

Hue=0

Saturation=53

Sharpness=0

Gamma=180

WhiteBalance=3447(Auto)

BacklightCompensation=0

Gain=0

Capture 15_02_2013 18_10_10 RAW STILL.bmp

Frame Divisor=1

Resolution=1280x960

Frame Rate (fps)=7.50

Colour Space / Compression=YUY2

Exposure=-10

Brightness=128

Contrast=32

Hue=0

Saturation=53

Sharpness=0

Gamma=180

WhiteBalance=3447(Auto)

BacklightCompensation=0

Gain=0

Capture 15_02_2013 18_14_54 RAW STILL.bmp

[This one had a 2x Barlow added]

Frame Divisor=1

Resolution=1280x960

Frame Rate (fps)=7.50

Colour Space / Compression=YUY2

Exposure=-9

Brightness=128

Contrast=32

Hue=0

Saturation=53

Sharpness=0

Gamma=180

WhiteBalance=3447(Auto)

BacklightCompensation=0

Gain=3

[JimD]

Doh!! Since I can't edit those bitmaps out of the post, I'll just have to repost the 3 images here.

[Steve Ward]

Start with the Gamma at zero and the Gain right up and adjust to suit.

[JimD]

Thanks for that, Steve - I'll give it go next clear sky.

[JamesF]

I'm not sure the gamma goes to zero on the Xbox cam, but as low as it goes is worth a try. It would be helpful to know what is the "no action" value but we're not the market these webcams are aimed at.

Before adjusting the gain I'd push the exposure time up a fair bit. Ideally you want to be getting ten or fifteen frames per second. Then pop up the histogram option in SharpCap and lift the gain until the histogram comes about three quarters of the way across. A lot of webcam imaging, especially when you're new to a camera, is trial and error, finding what works for your scope and your camera on whichever target you're interested in. I've produced some images of Jupiter and Mars that I'm really happy with using my SPC900, but now I've switched to an ASI120MC I'm in exactly the same place once again. Or nearly so, at least. This time around I know it can be made to work in the end

James

[JimD]

Thanks James. Unfortunately, it's not that easy to set the frame rate on the webcam - it seems to be linked to the resolution and can't be changed independantly.

Because I don't have a motor drive, my capture time is limited to about 30 seconds, which is the time it takes for Jupiter to cross the field of view (I daren't touch the scope to follow it, as the image bounces all over the place). It's recording about 250 frames in that time. Do you think that's enough for a meaningful stacking?

[inedible_hulk]

Hi Jim,

I started out with the Xbox cam, but I don't claim to be an expert. Two things you may wish to try.

1) If you reduce the resolution, you can up the frame rate. I was typically using 640x480 and a frame rate of 20 fps.

2) You might like to try a piece of software called PIPP, it's free. You can process your avi's for the best frames, but more importantly for you, it can batch process avi files and join them together. I'm not sure how well the latter will work for you, I've had mixed results, but it might be a good way to increase the number of frames available.

Clear skies

[JamesF]

1) If you reduce the resolution, you can up the frame rate. I was typically using 640x480 and a frame rate of 20 fps.

Definitely go for this. At the moment Jupiter is probably covering about 40 pixels per metre of effective focal length using the Xbox camera. Anything bigger than 640x480 is an awful lot of empty frame. And all the data for that larger frame has to be pushed over the USB connection which has limited bandwidth, so you can't transfer as many frames per second.

James

[JimD]

Definitely go for this. At the moment Jupiter is probably covering about 40 pixels per metre of effective focal length using the Xbox camera. Anything bigger than 640x480 is an awful lot of empty frame. And all the data for that larger frame has to be pushed over the USB connection which has limited bandwidth, so you can't transfer as many frames per second.

James

I don't quite get this bit.

Jupiter is quite small in the camera's field of view, but if I reduce the resolution it stays the same size, but has courser pixels. How does that improve the image?

[JamesF]

Some cameras with a sensor of (for example) 640x480 pixels can produce larger images, but they do so by interpolation and don't actually increase the amount of information in the image. From an imaging point of view there's no value in this -- if you're going to do it you might as well do so in processing, where time is less of an issue and you can therefore use better algorithms to do the job (and where image size is less of a problem).

Other cameras can produce an image smaller than their absolute resolution just by cropping off the edges of the image and using only a section from the sensor of the requested size.

Yet others can produce an image smaller than the size of the sensor by combining the data collected from a block of pixels and returning a single result as if it were one large pixel. This is known as "binning" (no idea why).

Some support two or even all three of these methods.

The Xbox camera fits into one of the first two categories. I think probably the first, but I'm not absolutely certain without checking.

James

[JimD]

I understand exactly what you're saying, James. I'm not sure exactly what this webcam is doing, though - I've refitted the webcam lens so that I can take daytime test shots, and it's without doubt capturing a higher resolution picture at 960 x 720 than it is at 640 x 480. The difference between 960 x 720 and highest resolution of 1280 x 960 is less discernable. The frame rate is set automatically, and jumps back to the auto-selected rate if you try to change it.

There seems to be a yellow cast over the picture at the lower resolution - don't know if that's significant or why it's happening.

Incidentally, the lowest gamma setting I have available in Sharpcap is 180. The range is 180 to 250.

I'm using Windows XP and the driver that was (automatically) installed is called "Microsoft XBox 360 Live Vision Security Method 3". Is that the same driver that others with this webcam are using?

[JamesF]

Not sure what I was using to be honest. I abandoned my Xbox cameras a while back and stuffed them away somewhere to try to clear a bit of space for things I do use. I'm not entirely sure where they are now.

James

[inedible_hulk]

Hi Jim,

These are the specs for the webcam.

http://en.wikipedia.org/wiki/Xbox_Live_Vision

I can't claim to understand it all either, but I have read that the increased resolution causes more 'noise' on the image. If I remember correctly this was linked to the webcam being only USB 1.0. I'm pretty sure I read that in Cliff's Xbox Webcam thread.

Any noise generated will be much more noticeable in astro images where you have much less light to play with. I never managed to get the colour right (see below), but it I learnt a lot from it before I moved onto the Phillips. I hope this helps

[inedible_hulk]

Hi Jim,

I'm on Windows 7 and the webcam shows up as the same, so I think you're using the right driver, either that or I'm not

If it helps, these were the settings I used last:

[Video Camera ]

Frame Divisor=1

Resolution=640x480

Frame Rate (fps)=20.00

Colour Space / Compression=YUY2

Exposure=-7

Brightness=93

Contrast=33

Hue=180

Saturation=56

Sharpness=52

Gamma=241

WhiteBalance=2770

BacklightCompensation=0

Gain=0

[JimD]

Thanks for posting your settings, IH, it's useful. I'll have to try this another day, though, as I haven't been able to get out tonight (despite having clear skies!)

[happy-kat]

I'm using Windows XP and the driver that was (automatically) installed is called "Microsoft XBox 360 Live Vision Security Method 3". Is that the same driver that others with this webcam are using?

My first xbox webcam works and is using that driver.

I bought a spare (2) and that uses a different driver and I get no image or communication with either of them.

I can't seem to get the driver that dows work available to the other webcam, the file on the Internet has a virus!

[xx Ash xx]

Hi Jim,

Like you, i'm a novice with the whole webcam photography subject.

But if you take a look at my earlier thread here http://stargazerslounge.com/topic/179015-second-go-at-jupiter/, you'll see the results that I got using the settings below

[Video Camera ]

Resolution=320x240

Frame Rate (fps)=10.00

Colour Space / Compression=MJPG

Exposure=-5

Brightness=128

Contrast=32

Hue=0

Saturation=53

Sharpness=0

Gamma=180

WhiteBalance=2995(Auto)

BacklightCompensation=0

Gain=147

If you take a look at post #8 on that thread, you can view the video that produced my image. I don't have a motor drive so I had to track Jupiter manually to the best of my ability. My total capture time was 5 minutes (al in one go) after which I put the avi through PIPP to align Jupiter and then through Registax 5 where I had to play around with the wavelets and also the contrast & brightness.

[happy-kat]

Thought I would post as have managed to get the webcam working, it is different to the other one, but it is now displaying an image. So hopefully all will now work.