Merged Discussion of femr's video data analysis

[femr2]

Oh dear - I just read thru the 'Seismic Events' thread and saw what prompted this latest offering from Femr2.

The prompt is simply the mention of the camera shake. Anything else you infer is your own responsibility.

I'd like to suggest that it is a meaningless correlation.

Perhaps, though there are numerous other events on the tower which also correlate.

Who knows what caused the camera to be disturbed - maybe, while filming this static shot for long periods of time, somebody moved over close to it, maybe they were eating a ham sandwich and got careless.

Indeed, thus the lookie-see. Perhaps it's another coincidence, but performing the traces has already been fruitful regardless. Am sure further clarity will be revealed.

Sure, you can use software to 'trace' the movements. So what? I fail to see what possible insight of value is ever going to come from this.

Really ? Well, for a start, identification of movement of WTC 1 ~9.5s in advance of release.

Again, would have thought you guys would be making a big deal with that metric, but hey

What I would love to see you do is measure some real controlled demolitions with your methods and determine what the acceleration plots look like.

What on earth for ? You already know my opinion on the descent mechanism (ROOSD) which is entirely gravity driven. But I have no problem doing so. If you point me to a specific decent quality video with a static camera, preferably as little perspective distortion as possible, I'll sort you out with the trace data.

Then we could use them to compare to the 3 WTC collapses...especially if you'd include WTC 2, which seems to be permanently off the radar with the Szamboti/Major_Tom crowd.

Well 1 & 7 are already available, 2 can and will be done at some point.

 

[DGM]

Femr2:
I know it's most likely common knowledge to you but, where exactly was the camera mounted and what was it mounted on?

 

[femr2]

Femr2:
I know it's most likely common knowledge to you but, where exactly was the camera mounted and what was it mounted on?

Don't have the info to hand, but...Camera location was 145 Avenue of the Americas ~1.5km I think.

Will dig out the more precise position dms and elevation, and camera details. Not sure If I've seen confirmation of exactly what mount, but certainly tripod.

Why ?

 

[DGM]

Don't have the info to hand, but...Camera location was 145 Avenue of the Americas ~1.5km I think.

Will dig out the more precise position dms and elevation, and camera details. Not sure If I've seen confirmation of exactly what mount, but certainly tripod.

Why ?

I think that would be number one if you are trying to connect ground movement to the camera.

 

[femr2]

I think that would be number one if you are trying to connect ground movement to the camera.

The primary purpose is to determine early building movements. It's simply observation that all movements seem to begin at that point in time.

Whether the camera shake is coincidental or not makes no difference to the building movement observations.

I'll certainly be posting additional detail on events on the tower which also occur at that time. These events include ejection of pieces of the tower, pressure pulses resulting in rapid smoke volume increase, westerly features moving eastwards whilst easterly features moving westwards...that sort of thing.

 

[BigAl]

The primary purpose is to determine early building movements. It's simply observation that all movements seem to begin at that point in time.

Whether the camera shake is coincidental or not makes no difference to the building movement observations.

I'll certainly be posting additional detail on events on the tower which also occur at that time. These events include ejection of pieces of the tower, pressure pulses resulting in rapid smoke volume increase, westerly features moving eastwards whilst easterly features moving westwards...that sort of thing.

As of 2010 the only game changer will be the simultaneous sound of demolition explosions on video tapes from several cameras.

Let us know if you come across them.

 

[pgimeno]

No. There are so many grey-variables in latching the seismic data to a specific time that it becomes very unwieldy, especially with poor quality graphs being the only data I've seen.

If the numeric seismic data is available, I'd be happy to do so. (or high fidelity graphs I suppose)

Does any other footage show any camera shake?

Does it show significant shake during the period where the generation of the most prominent seismic signals is expected?

 

[femr2]

Does any other footage show any camera shake?

Not sure. Will be checking other fixed camera footage.

Does it show significant shake during the period where the generation of the most prominent seismic signals is expected?

 

[pgimeno]

Care to explain what the graph is, label the axes and put times, please, or alternatively answer the question directly instead of with an unlabeled image?

 

[femr2]

Care to explain what the graph is, label the axes and put times, please, or alternatively answer the question directly instead of with an unlabeled image?

The graph is vertical camera movement (trace of near camera static point).

Upper field only, 29.97fps.

Vertical axis is pixels, horizontal axis is frames.

First period of high amplitude change is camera wobble ~9.5s in advance of release (for ~5s).

Release around frame 1100.

Second longer period is descent and wotnot (for ~30s).

Sorry, I'm far too used to the various trace shapes.

 

[pgimeno]

Thanks. Could you please pass the data through a high-pass filter, to remove the zigzag and better appreciate the magnitude of the shaking, maintaining the axes?

 

[femr2]

Thanks. Could you please pass the data through a high-pass filter, to remove the zigzag and better appreciate the magnitude of the shaking, maintaining the axes?

 

[DGM]

Curious question. Any idea what a trace of a video taken from a camera with video stabilization would look like if someone bumped the tripod?

 

[alienentity]

. If you point me to a specific decent quality video with a static camera, preferably as little perspective distortion as possible, I'll sort you out with the trace data.


Well 1 & 7 are already available, 2 can and will be done at some point.

How 'bout this one from your channel?
http://www.youtube.com/watch?v=c5iFW89tw2k

or this one from mine?
http://www.youtube.com/watch?v=3GNhEpHfgfI

or this one

http://www.youtube.com/watch?v=kVlC9rYU-gs

 

[femr2]

Curious question. Any idea what a trace of a video taken from a camera with video stabilization would look like if someone bumped the tripod?

I don't have a tripod I'm afraid, but if someone uploads an appropriate video, I'll trace it.

I'd suggest, however, that the period of shake would be much less than the sustained 5s of the shake being plotted.

I'll also emphasise that the primary focus is the movement on the tower beginning at the same time, rather than the shake itself.

 

[femr2]

How 'bout this one from your channel?
http://www.youtube.com/watch?v=c5iFW89tw2k

or this one from mine?
http://www.youtube.com/watch?v=3GNhEpHfgfI

Neither are great, but can do. (Third one won't work for some reason)

Not sure of the point in comparing verinage, but you're welcome to the data to do with as you will.

Before I begin though, what are you intending to look at ?

If velocities and acceleration (or indeed anything in real world units), then I'll need you to source building size metrics first, such that pixel based movements can be translated.

 

[DGM]

I don't have a tripod I'm afraid, but if someone uploads an appropriate video, I'll trace it.

I'd suggest, however, that the period of shake would be much less than the sustained 5s of the shake being plotted.

I'll also emphasise that the primary focus is the movement on the tower beginning at the same time, rather than the shake itself.

I would think you'd want to eliminate all the trivial explanations before you tried to connect the two. Coincidences happen all the time.

 

[carlitos]

I would think you'd want to eliminate all the trivial explanations before you tried to connect the two. Coincidences happen all the time.

Obviously, unless you are trying to prove some type of conspiracy.

I joined imdb pro, and neither of the Naudet brothers has contact information available. Before I spend much effort finding them, has anyone here previously asked them about this particular camera shake?

Thanks!

 

[femr2]

I joined imdb pro, and neither of the Naudet brothers has contact information available. Before I spend much effort finding them, has anyone here previously asked them about this particular camera shake?

Thanks!

Don't think you'll get too far with the Naudet's...the video was shot by Etienne Sauret.


Still no mention from folk about identified movement ~9.5s in advance of release. Quite a surprise.

 

[femr2]

I would think you'd want to eliminate all the trivial explanations before you tried to connect the two.

Again, perhaps it was knocked or due to the helicopter, perhaps not. Due to the behaviour of the shake, I doubt it was either of those, but regardless, the point is that numerous other events occur on the tower at the same time, all of which either suggest shaking or the earliest moments of an elongated initiation process relatively far in advance of release.