Webcam on the road

[Bill]

I use a webcam so that I can hook up trailers easily to my Disco without needing anyone to say left hand down a bit etc. It works excellently for this. The camera is a no-name tiny thing that has the lens at right angles to the thin gooseneck type stalk that I gaffer tape to the Disco's gutter. It uses "HP Basic Starter" camera drivers into the software that came with my Vista notebook. This sits on the front passenger seat of the diesel auto car.

So I thought I'd try gaffering the camera to face forward and record a video of the journey home, possibly to edit and send to friends abroad so they can marvel at our potholes and so on.

It worked OK, with only a few dropped frames, but when I play it back it almost makes me seasick to watch because of what looks like the shimmering of the picture with what look like variable speed rolling hum bars. These relate to engine speed, and I assume must be some sort of magnetic field from the car electrics. The laptop is on batteries and there is only the usb webcam connected.

Has anyone any experience of this sort of thing before I try to bodge a ferrite lump onto the usb cable or buy a steel box for the laptop?

[Adrian C]

I'd gently strip apart the camera and have a look. Particulary at timing components and supply filtering (or lack of). Maybe a heavy part (electrolytic?) has come loose under the vibration of a badly soldered joint, and there is some analogue circuitry on the camera board that is now getting upset.

By the time the streaming signal has left the camera, it's in the digital domain. Adding filters here or to the laptop won't do much, unless added to the individual supply lines in the USB connector. However, noise here should be filtered out by whatever regulator is used in the camera.

For here, is were I think something's probably broke ...

[John M]

Can't help much I'm afraid, only experience is my daughter taking a similar video using one of those really small spy cameras about the size of a biro top which she clipped onto the sun visor. results were good but only seen when she got home (ie she didn't use it as a web-cam)

[Dave Liquorice]

Might be vibration related if the camera was mounted firmly to the body work. Only a small vibration upseting the interlace and/or the video compression encoding. Do you get the same effect if a passengerhand holds the camera?

It could be induction of some sort from the engine, try a different place, does it change?

[John Williamson]

Can you post a short sample anywhere?

[djc]

Has the camera got some form of electronic image stabilisation? I've recently tried riding a bike with a camera mounted on the frame. The shimering effect maybe the stabilisation not being able to keep up.

[Tim Lamb]

Is the Disco diesel? If so unlikely to be ignition rf.

regards

>

[Bill]

In message , John Williamson writes

Unless anyone can suggest a simple place to post a small temporary video file, I'd have to investigate this with a son who has posted to YouTube and elsewhere, meanwhile thanks to all for the replies.

I've been fiddling around again with the engine running and am beginning to think it might be vibration rather than induction. I can mount it in the gutter and see shimmying come and go, but if I hand hold it anywhere close to the same position, the picture is steady.

I don't think it's an electronic fault in the camera, and it's so small that I can't see a repair being possible. It's good because it can be focussed and yet is very small and easy to fit on the car. Tomorrow I'll try to fit in a run with it attached with a shorter length of free gooseneck and a bit of foam where it rests on the gutter.

I don't think the camera has any image stabilisation - it won't have been expensive, but looking again at the video, there are more dropped frames or glitches than I first thought. The laptop is a 2GHz dual core, and the video being captured is only 640 x 480, so I'd thought it ought to be able to keep up. I did change to what I thought was a free version of "Debut" software, which seemed to work better, but when I came to use it this time it said my trial had expired.

I realise that this has all the signs of a cheapskate trying to do something for nothing, but it would be nice to get it working.

[Dave Liquorice]

The OP said it was... Doesn't stop there being all manner of other electrical signals flying about. If it's a TD5 it's completely "fly by wire", no direct physical connection between the controls and the engine.

Looks like it's a vibration issue upseting the camera image generation/processing somehow. Bit of soft sponge in the mount may well cure it.

I'm all for cheap when it's just for fun. If serious you'd invest in one of those always recording, GPS logging, cameras and associated software and maybe a good mount. I quite like videos done from vehicles with the camera in a stabilsed mount so the relationship between the view outside and the camera remains constant with vehicle swinging about all over the place.

[Jon Fairbairn]

If it hasn't got image stabilisation, you are likely to get this sort of effect from vibration alone if it has a strong vertical component near the frame rate, simply because the position of the image oscillates and strobes with the rate at which frames are taken. You can get a similar optical effect while watching TV if you growl at about 50Hz (or watch with a shaver pressed against your chin or something).

[Bill]

Well, thanks to all. It must have been vibration.

I've just done a run around the neighbourhood with the camera mounted through a small length of foam pipe insulation and it's steady.

In the process I've discovered that something has chewed through the polythene pack and started to eat the pipe insulation, and the software that said the registration had expired yesterday now says it is the free edition.

Now to improvise a less crude camera mount.