Extracting and animating data from the "UFO Video Captured By Homeland Security"

a reply to: IsaacKoi

Yes, I figured that is why it was converted to a PDF. I thought writing an app would be quicker because you can scan just the area you want, and input it directly to a spreadsheet. Also, it provides more control over the OCR settings. It may still be a doable option, just with a bit more time, which I don't have much of.

originally posted by: WeAre0ne
I thought writing an app would be quicker because you can scan just the area you want, and input it directly to a spreadsheet.

Interesting approach.

I've have to keep that trick in mind for future projects.

Exceptional work as usual Isaac. I won't be able to participate much until the end of next week, if anything is left then I'll be glad to help.

a reply to: WeAre0ne
I had the same idea about using OCR. I used RoboRealm and didn't have much luck either. I think what you have to do is run the video through some filters to bring out the text more. Definitely a time suckage.

Isaac: There may be some other tools within RoboRealm that may be useful. Its vision software for robotics.

originally posted by: ZetaRediculian
I had the same idea about using OCR. I used RoboRealm and didn't have much luck either. I think what you have to do is run the video through some filters to bring out the text more. Definitely a time suckage.

Mmm. Given the apparent problems with OCR (and the probable need to check any output anyway), at the moment I think that continuing with the boring (but not THAT time consuming...) job of entering the figures manually is probably the best option.

As I mentioned above, since the numbers do not change with each frame it is possible to scroll through the frames and cut and paste the same number for a column into a number of rows (sometimes hundreds of rows at a time).

I'll probably have time to do more of the columns at the weekend if they remain to be done.

I think we should be ready to focus on the animation aspect relatively soon.

a reply to: IsaacKoi

Mmm. Given the apparent problems with OCR (and the probable need to check any output anyway), at the moment I think that continuing with the boring (but not THAT time consuming...) job of entering the figures manually is probably the best option.

Yeah, I'm faced with that decision regularly, manual entry or code. Often I opt for manual entry. I may still play with it, mostly because I haven't used the OCR tool yet and this looks like good practice. I am keeping an eye on this and will let you know if I can come up with anything useful.

originally posted by: ZetaRediculian
I haven't used the OCR tool yet and this looks like good practice. I am keeping an eye on this and will let you know if I can come up with anything useful.

Sounds good. Even if we get this particular task done by manual entry of the remaining columns, knowing about improved OCR techniques will doubtless be useful down the line.

Link to link in another post 404's. Picture looks like a bacterium through a microscope. Homeland Security ensigns can be Photoshop'd into anything to make them look "top secret"...

Summation: 5% Ufology; 95% Ufoism.

originally posted by: PresidentHeston
Link to link in another post 404's. Picture looks like a bacterium through a microscope. Homeland Security ensigns can be Photoshop'd into anything to make them look "top secret"...

Summation: 5% Ufology; 95% Ufoism.

Still better than 100% trolling and religious quakery

Let's try to keep the focus on the challenges I mentioned in the OP, rather than attempting to express conclusions at this stage:

originally posted by: IsaacKoi
Challenge 1 : Extracting data from the video into a spreadsheet (partially completed)

Challenge 2 : Plotting and animating some of that data (whether using After Effects, Cinem4D or otherwise) on a map or maps (whether sourced from Google Earth or otherwise).

originally posted by: chunder
An anomalous event, some hard data, a detailed report and now peer review

'Peer review' ?
If you think this is peer review then you are sadly mistaken.
Lol.

S&F
Tons of work here and I look forward to seeing your results!

originally posted by: DogMeat
I look forward to seeing your results!

I look forward to seeing them as well.

a reply to: IsaacKoi

Give me a column to do. I will have some free time tomorrow.

man what an awesome topic. not here to help, just to say thank you

originally posted by: raymundoko
Give me a column to do. I will have some free time tomorrow.

Great.

Next column up is AH, so I'll take you as having assigned yourself that one.



So far, the list of volunteers is:

Column AD : Excallibacca

Column AH : Raymundoko


Any problems, just give me a shout.

Neat you found what data you did on the camera. Too bad the operator's manual (or TECHMAN equivalent) doesn't seem to be among the .pdf files. But sales brochures do give a bit of spec that lets you figure out the FOV. If I had any guesses based on what little the spec sheet says, auto and default probably have to do with stabilization mode and whether or not IR is overlayed with optical. (One of those neat features built-in to that particular camera.)

As for making a rudementary animation reconstructing this event? Why not have something in 3D using Blender? Wouldn't be too hard to create a .blend with some basic assets and have it available for sharing. At the beginning it wouldn't have to be too pretty, just an idea of where the plane, UFO are relative to each other, and where the camera is mounted and perhaps a highlighted volume indicative of the camera's field of view.

Like some people have mentioned, I'd think the object was going pretty straight. Moving and jostling around has to do with the operator or automated tracking losing lock.

originally posted by: IsaacKoi
Let's try to keep the focus on the challenges I mentioned in the OP, rather than attempting to express conclusions at this stage:

originally posted by: IsaacKoi
Challenge 1 : Extracting data from the video into a spreadsheet (partially completed)

Challenge 2 : Plotting and animating some of that data (whether using After Effects, Cinem4D or otherwise) on a map or maps (whether sourced from Google Earth or otherwise).

I hugely appreciate your scientific and objective approach to analyzing the data.

My question is that investigation usually has the aim of proving or disproving a premise.

Not trolling...but what is there in this imagery that warrants further investigation?

What are the anomalous factors that warrant deeper analysis?

Put another way, that risks backlash...what unusual data or imagery makes you believe that this might not be a simply a Mylar balloon spinning at the whims of the wind captured by semi-primitive camera systems designed to detect more obvious Arial phenomena like aircraft?

a reply to: Indigo5

Because a 161 page report came out and was gobbled up by the UFO community, but it was chalk full of errors, many of which ruled out a balloon or bird only due to the erroneous assumptions and shoddy investigative work done in the report.

This peer review will either confirm that report, or show the errors.

originally posted by: Indigo5
what unusual data or imagery makes you believe that this might not be a simply a Mylar balloon spinning at the whims of the wind captured by semi-primitive camera systems designed to detect more obvious Arial phenomena like aircraft?

I don't recall saying that I believe that this might not be simply a Mylar balloon...



You may not be familiar with the recent background. No reason why you should be, nor did I go into detail on it in my OP. Another group of researchers ("SCU") posted a long report which categorically ruled out a balloon. (I linked to that report in my OP). Shortly after the report was released, I briefly voiced some doubts about its conclusions - offending one of the authors of the report (although others of them have remained very friendly and cooperated with my requests).

The work I'm doing (with a bit of reluctance...) is aimed at critically evaluating the SCU report.

Basically I think that taking a few steps (which should not be incredibly difficult, even if some of them are a more than a little boring) may well conclusively establish the position, going a bit beyond some of the rather sterile debates that sometimes occur within ufology.