- This topic has 6 replies, 6 voices, and was last updated 4 years, 5 months ago by
.
-
Posts
-
One consequence of the lockdown is that I’ve had more time for toy projects lately, including reviving my old project to set up meteor cameras around a Raspberry-Pi-based platform: https://pigazing.dcford.org.uk/
Out of curiosity, how often do people refocus their meteor cameras, and how do they go about doing it? It’s a bit of a pain, as it either involves bringing the camera indoors, or taking a portable monitor up a step-ladder.
My cameras seem to stay in reasonable focus for a week or three, but in recent days they’ve drifted significantly out of focus, which I guess is a consequence of the warm weather.
Hasn’t been a problem for me Dominic, to be honest.
Have two cameras, both 6mm f0.75 lenses. Both keep their focus.
Out of curiosity, how do you get the Pi to detect the Watec camera?
Ditto for me. I haven’t refocused mine for years (3.8mm, f/0.8). My main problem is spiders. Now that planes are mostly gone they are a major source of false triggers on my cams. I have a brush on a long pole for dealing with them but they don’t stay away for long.
Hi Dominic,
Same for me too – haven’t had to refocus my cameras for years.
The lenses I use have a grub-screw on the rotating bezel which allows the focusing ring to be locked in position once focus has been achieved. In the absence of a screw you could use a small pice of masking tape to hold the focussing ring in the same position relative to the lense barrel.
As for how I achieve critical focus, I point the camera near the zenith, choosing a patch of sky that contains a combination of bright and faint stars and adjust the bezel until the faint stars become visible (when you are not in focus, the light from the faint stars is smeared out, rendering them invisible against the background sky).
Agree with Nick – spiders are a problem this time of year, as well as pollen (distinct lack of rain recently to wash the front window clean).
Best regards
William
Hi Dominic,
As William commented, my lenses also have a grub screw to secure the focus of the lens. As you’ll have experienced, focussing is quite a delicate operation with most video camera/lens combinations.
My Leeds_SE camera sometimes drifts out of focus during the seasons. Often this is an acceptable amount of soft focus, then just as I’m about to shin up the drainpipe it moves back into hard focus again.
I haven’t tried the masking tape technique. I suspected its expansion and contraction might cause some focus shift. I suppose it depends on the tightness of the rotating bezel.
Cheers,
Alex.
Thanks – that’s interesting.
My lens doesn’t have a grub screw to hold the focus, so perhaps a bit of tape or glue is the way to go. The quality of focus doesn’t seem to strongly affect the number of meteors I record, so I may be being too picky. I guess meteors are moving so quick that they’re inevitably spread over many pixels, so having a soft focus doesn’t make such a difference. The one benefit to a really sharp focus is that I see a lot more stars, which is handy for calibration.
I use an easycap usb dongle to connect the Watec 902H2 Ultimate to the RPi. See the photo below. The whole setup fits inside a Genie TPH2000 enclosure, powered via power-over-ethernet so only one cable is needed. A custom Pi Hat converts the 48V supply from the PoE into a 5V supply for the RPi and a 12V supply for the camera. There’s a transistor to allow the RPi to turn the camera on/off via one of the GPIO control lines. Previously I was using a RPi model 2, which had issues with dropping frames. But I’m now using a RPi model 4, which is a pretty fast machine. Very occasionally frames still get dropped, possibly due to USB errors, but 99% of the time it’s fine.
At some point I’d like to get my software to produce output in a compatible format with UFO Capture, so that we can share data. At the moment my biggest concern is calibrating the pointing, though. I’m doing a polynomial fit to the radial distortion in the lens, and then using astrometry.net to determine the pointing of each image. I periodically take one-minute exposures through the night for this purpose. I then take the median of all the pointings determined each night to arrive at a single (alt, az) estimate for each camera each night.
It works up to a point… but I’m still seeing ~5 pixel offsets in the fitting a lot of the time… which isn’t very good.
I am not too sure how you “polynomial fit to the radial distortion in the lens”. Is it a one-off calibration done back in the distant past by, say, imaging a sheet of graph paper?
Alternatively, do you apply a rough calibration then iterate (plate-solve, polynomial-fit) until convergence?
Further, is the calibration polynomial assumed to depend only on the radial coordinate? If so, have you considered a full (x,y) polynomial fit? The distortion *should* be angle-independent but if you haven’t located the coordinates of the centre pixel (centre of the lens, not the centre of the detector) accurately enough you are likely to have problems.
- You must be logged in to reply to this topic.