Check Six!

In a previous article, said:
That's how I did this video, with an earlier, lower-resolution camera:

http://www.youtube.com/watch?v=ubDOG4E_pXs

That's some weird compression artifact that causes that strange black
stripey stuff in the prop disk.


--
Paul Tomblin http://blog.xcski.com/
So logically, if she weighs the same as a duck, she's made of wood, and
therefore a witch.

On Tue, 20 Feb 2007 00:58:52 +0000 (UTC), (Paul
Tomblin) wrote:

In a previous article, said:
That's how I did this video, with an earlier, lower-resolution camera:

http://www.youtube.com/watch?v=ubDOG4E_pXs

That's some weird compression artifact that causes that strange black
stripey stuff in the prop disk.

Slow frame rate, I think. The camera that took the video on my original movie
is low resolution and only has about 10 frames per second. The new one is
nearly full NTSC resolution with 27 FPS. Same manufacturer, just ~2 years'
improvement in technology (the cameras both sold for about $90.

As Andy noted, even the new camera's sampling rate of the focal plane tends to
show the propeller of the Boredom Fighter as an arc. Certainly not as bad as
the rising/falling ~ of the old camera, but still noticeable.

I just too cheap to spend money for a real video camera, that's all. :-)

Ron Wanttaja

That's some weird compression artifact...

Slow frame rate, I think.

Neither. The "shutter" speed actually looks to be around an effective 1/5000 of a second.
Normal video is 1/60. The distortion of the prop is not due to compression, but due to
how the image is scanned off of the sensor chip... sequentially, row by row, bottom to
top. That is, if it takes 1/5000 of a second to read each row of pixels, then multiply
that by 240 to 480 rows and you get 10 to 20 full frame images every second.

The position of the prop changes by a couple degrees by the time the next row of pixels up
on the chip is read by the camera's firmware. The result is the shearing, bending, and
floating pieces of the blades when the whole image is assembled. Very disconcerting since
we are all too accustomed to our human vision having analog motion blur. This visual
distortion is called Temporal Aliasing (digitally sampled 'stair-stepping' of time).

So Ron. What I'd like to see is how you mounted the camera. There was virtually no high
frequency vibration in that mount... very rigid. Impressive. Plus being that these
cheapo digital cameras are tapeless (record directly to flash memory as an mpeg 4 file),
there's no breakup of the picture due to tape-to-record head gaps from a vibrating tape.
Going to a better, high-end camcorder (even the best mini-DV), your picture is no doubt
going to go in the toilet of digital dropout and break up unless you use an external video
recorder appropriately vibration/shock mounted in the cockpit to prevent tape vibration
that separates it from the rotating head drum.

Dean Scott

--
Posted via a free Usenet account from http://www.teranews.com

On Wed, 21 Feb 2007 10:03:46 -0500, "Dean A. Scott"
wrote:


That's some weird compression artifact...

Slow frame rate, I think.

Neither. The "shutter" speed actually looks to be around an effective 1/5000 of a second.
Normal video is 1/60. The distortion of the prop is not due to compression, but due to
how the image is scanned off of the sensor chip... sequentially, row by row, bottom to
top. That is, if it takes 1/5000 of a second to read each row of pixels, then multiply
that by 240 to 480 rows and you get 10 to 20 full frame images every second.

That makes sense. IIRC, there is a type of still-camera shutter that works
similarly, and you sometimes see the same kind of "arcing" of moving propellers.

So Ron. What I'd like to see is how you mounted the camera. There was virtually no high
frequency vibration in that mount... very rigid. Impressive.

Beginner's luck. The cameras were basically mounted on an extension of a big
aluminum angle, which was clamped to the axle by hose clamps. I put piece of
inner-tube rubber under the angle/clamps to protect the paint of the axle.
Details, including photos, on:

http://www.bowersflybaby.com/pix/video.html

The one problem I have is a low-frequency shake of the image. If you watch the
ground in the distance, you can see it's shaking up and down at about a 1/2 to 1
Hz rate. Both cameras do it, the newer camera has electronic stabilization, and
it didn't seem to help.

My guess is that it's that bit of rubber that's protecting the axle...I think
the mount is "bouncing" a bit. The effect is not really visible with the old
camera when it was mounted atop the tail, where there isn't any rubber involved.

Gonna try to protect the axle with nylon or wood, instead.

Ron Wanttaja

"Ron Wanttaja" wrote

My guess is that it's that bit of rubber that's protecting the axle...I
think
the mount is "bouncing" a bit. The effect is not really visible with the
old
camera when it was mounted atop the tail, where there isn't any rubber
involved.

Gonna try to protect the axle with nylon or wood, instead.

My guess is that it is not the rubber, but instead, the axle flexing.

Try mounting the camera on the axle over far to one side, right next to the
axle support, and see if you still get the "bounce." Use the rubber and
everything else just the same, for a control.
--
Jim in NC