|
|
|
Thread Tools | Display Modes |
#11
|
|||
|
|||
Why .avi format ?
Michael McCulloch wrote:
On Sun, 27 Nov 2005 18:28:20 -0500, Tom Rauschenbach wrote: From reading this newsgroup I understand that .avu format seems to be the conventional file format for digital imaging. I usually associate that format with motion pictures. Why is that the format used for time exposures through a telescope ? The AVI format is only used for solar system imaging where the object is not dim (e.g. the planets and the moon), This is not strictly correct. Check the Yahoo group QCUIAG for the popular modified webcams which are quite capable of long exposures to image also deep sky objects. Or see my web page :-) the exposure is 1/10 sec or so, and the goal is to get numerous images that 'freeze' moments of good seeing from which to pick the best for stacking to increase contrast and decrease image noise. True for planetary/solar/lunar imaging. The exposure is usually much shorter than 1/10 sec though. Time exposure astro imagers do not use AVI. This is not true. I have several long exposure modified webcams and they all produce AVI files. I also have a higher end camera producing FITS. Clear skies Carsten A. Arnholm http://arnholm.org/ N59.776 E10.457 |
#12
|
|||
|
|||
Why .avi format ?
On 28 Nov 2005 04:40:37 -0800, "
wrote: This is functionally equivalent to high order adaptive optics. Not really. High order adaptive optics allow to correct the exit pupil phase differences while select-and-stack simply rejects those that do not qualify. Which, in terms of the final image, is substantially similar. Sure, discarding bad images is equivalent to AO with very poor QE, but all the same, I consider such processing to effectively be post-processed AO. Readout noise makes for a stiff noise penalty... This depends on the read-out noise. The lower it is the less different the 2 techniques are. Of course, but realistically all cameras now being used are high readout noise. Webcams are particularly poor in this respect. Once you start imaging DSOs, you need long exposures- many minutes is usually required to maximize S/N. Not really. I've been imaging DSOs for the past 4 years without ever taking exposures longer than 120s, 45s to 60s being the most common durations. With low read-out noise camera there are benefits in using short exposures. Unless you are imaging under horribly light polluted skies, it is likely that you are not optimizing your S/N with such short exposures. That isn't the same as saying that you aren't getting good results, however- just that they could be better. There can be benefits to short exposures, particularly if your equipment isn't good enough to support accurate tracking for long periods. I have an experimental camera with essentially zero readout noise. You can do some very cool things with it, such as building the final image dynamically from subsecond exposures. The actual image is used for guiding, and the current sum image is always available, so the exposure can be run until the desired goal is visible. Focus can be monitored and controlled throughout the exposure, and bad frames can be rejected on the fly. Cool stuff, but these sensors don't seem to be under very aggressive development (although I guess they are being developed for military applications, since they perform better than the best intensifier-based night vision equipment). _________________________________________________ Chris L Peterson Cloudbait Observatory http://www.cloudbait.com |
#14
|
|||
|
|||
Why .avi format ?
Chris L Peterson wrote: On 28 Nov 2005 04:40:37 -0800, " wrote: .... Readout noise makes for a stiff noise penalty... This depends on the read-out noise. The lower it is the less different the 2 techniques are. Of course, but realistically all cameras now being used are high readout noise. Webcams are particularly poor in this respect. True. But webcams are really the lower end of the video imaging spectrum. Once you start imaging DSOs, you need long exposures- many minutes is usually required to maximize S/N. Not really. I've been imaging DSOs for the past 4 years without ever taking exposures longer than 120s, 45s to 60s being the most common durations. With low read-out noise camera there are benefits in using short exposures. Unless you are imaging under horribly light polluted skies, it is likely that you are not optimizing your S/N with such short exposures. That isn't the same as saying that you aren't getting good results, however- just that they could be better. I opine that while it is true that longer exposures will produce higher S/N images (with the same length of time) it is not true that they are going to be necessarily better. If the subject being imaged is bright (relative to the scope/camera combination at hand) shorter exposures (30s-60s) will go a great length in improving resolution, if you haven't got a tip-tilt AO. There is also the problem with oversaturation of bright stars in the field that can be worked around. Of course, as you're going to discard all not-too-good frames, to build up a an image with decent S/N might entail doubling the effective exposure time. BTW, my skies are pretty good (when is not cloudy, that is), at about 5.5 ZLM. There can be benefits to short exposures, particularly if your equipment isn't good enough to support accurate tracking for long periods. I have an experimental camera with essentially zero readout noise. You can do some very cool things with it, such as building the final image dynamically from subsecond exposures. The actual image is used for guiding, and the current sum image is always available, so the exposure can be run until the desired goal is visible. Focus can be monitored and controlled throughout the exposure, and bad frames can be rejected on the fly. Cool stuff, but these sensors don't seem to be under very aggressive development (although I guess they are being developed for military applications, since they perform better than the best intensifier-based night vision equipment). As you said, cool stuff. Are they available commercially? Best Andrea T. |
#15
|
|||
|
|||
Why .avi format ?
David Nakamoto wrote: For even Jupiter (Saturn normally does not present any cloud features that show up in amateur imaging) a few minutes of video can acquire a thousand images or more, and a few minutes is not long enough to show motion of the cloud features, at least at image sizes where Jupiter is perhaps a quarter to a third the width of the video, so I don't see this as a major reason to not go video as I think you're implying. First, amateurs have taken images showing small transient features (w.o. but not only) on Saturn since few years ago, even with "small" apertures (I did it with a 8" MN/MCT, see http://www.geocities.com/andreatax/2...opposition.htm or http://www.geocities.com/andreatax/2...opposition.htm). Secondly, the intrinsic limit of the length of a video on a planet like Jupiter is much less then you might think of, 120s-160s is about as good as it gets at its maximum angular diameter with a 10". I'm assuming a sampling of 4 pixels per airy disc diameter. Saturn can even be worse than that as it is far dimmer than Jupiter, although the length of the video recording can be 2.5x that of jupiter. To build a high resolution, good quality image of saturn you really need thousands of good images and that only happens when the seeing is extremely good. It's true that adaptive optics can produce more consistent, sharper images, but for those that don't have the equipment to do it, video is a good way to go. As for the readout noise, are you sure you're not referring to sensor or thermal noise? Thermal noise is irrelevant. We're talking about readout noise. Andrea T. |
#16
|
|||
|
|||
Why .avi format ?
On 29 Nov 2005 00:38:24 -0800, "
wrote: True. But webcams are really the lower end of the video imaging spectrum. Yes, but even most high quality video cameras have high readout noise. It is intrinsic to the CCD sensors available, and cheap electronics just adds to it. I opine that while it is true that longer exposures will produce higher S/N images (with the same length of time) it is not true that they are going to be necessarily better. If the subject being imaged is bright (relative to the scope/camera combination at hand) shorter exposures (30s-60s) will go a great length in improving resolution, if you haven't got a tip-tilt AO. Interesting. Different people report different results here, so I think it has a lot to do with the specific details of your local seeing. I've made thousands of seeing measurements, looking at the frequency spectrum of the shift, and for me, the short exposure advantage ends at about one second. That is, I'll see reduced star sizes up to about one second, but no difference between star sizes for longer exposures. There is also the problem with oversaturation of bright stars in the field that can be worked around. This is the usual reason I will choose to shoot short exposures. It costs me in noise, but keeps all my data linear. Of course, a decision like this depends also on the ultimate intent in collecting the data. As you said, cool stuff. Are they available commercially? Not really. A few very specialized (non astronomical) cameras using the sensors are available, but are extremely expensive. The sensors themselves are available (or were a few years ago) as engineering samples, but you need to make your own camera around them. The one I have has lots of problems- dead columns, hot pixels, and the like, so I don't use it for more than testing. If they ever get the bugs worked out of the technology, and these sensors become common, it will revolutionize astroimaging. _________________________________________________ Chris L Peterson Cloudbait Observatory http://www.cloudbait.com |
#17
|
|||
|
|||
Why .avi format ?
Chris L Peterson wrote: On 29 Nov 2005 00:38:24 -0800, " wrote: True. But webcams are really the lower end of the video imaging spectrum. Yes, but even most high quality video cameras have high readout noise. It is intrinsic to the CCD sensors available, and cheap electronics just adds to it. You'll be amazed on how low it is for very high end video cameras. They use mainstream CCD that are used also by cooled CCD cameras manufacturers. And if you think what the dedicated CCD they use for "fast" AO can do the best has yet to come... Reg's Andrea T. |
#18
|
|||
|
|||
Why .avi format ?
On 30 Nov 2005 06:37:22 -0800, "
wrote: You'll be amazed on how low it is for very high end video cameras. They use mainstream CCD that are used also by cooled CCD cameras manufacturers. All of which have high intrinsic readout noise (typically 8-15 electrons). And this readout noise is worse when the chip is clocked fast- pretty much a requirement for video. Especially at high speeds, it is extremely difficult to design a CDS amplifier that doesn't add yet more noise. Since this is the only thing under the control of the camera designer, I guess it's where most of the effort goes with very high end video. But still, they can't get better than the sensor specs, which is why all video cameras- regardless of price- are readout noise limited when used on astronomical targets. _________________________________________________ Chris L Peterson Cloudbait Observatory http://www.cloudbait.com |
#19
|
|||
|
|||
Why .avi format ?
Well, I wouldn't call 8 e- "high", expecially in planetary imaging.
Besides I was think to something in the range of 2-3 e-. Andrea T. |
#20
|
|||
|
|||
Why .avi format ?
David Nakamoto wrote:
AVI is a videofile format. Rather, AVI is a container format. Inside an AVI file, the movie can be coded with any of a variety of codecs, typically DivX, XviD or some of Intel's old Indeo codecs. Recently, however I have seen a lot of AVI file that are actually MPEG4 or even h.264 (HD-video), which both run natively in QuickTime 7. Webcam movies are most likely Indeo 5.0 or DivX. -- I recommend Macs to my friends, and Windows machines to those whom I don't mind billing by the hour |
|
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Medium Format "Piggyback" support/drive/tracker- suggestions? | rcyoung | Amateur Astronomy | 2 | September 22nd 05 07:59 PM |
Program to Convert Image to SBIG Format | matt | Amateur Astronomy | 6 | May 23rd 04 06:56 AM |
Tak FSQ106 and medium format camera | Rockett Crawford | Amateur Astronomy | 2 | February 18th 04 06:42 AM |
./setiathome: Exec format error. Binary file not executable. | Rollo | SETI | 2 | January 5th 04 10:32 PM |
Medium format film question | RalphMW | Amateur Astronomy | 3 | October 8th 03 03:19 AM |