Utah Drone video of UFO [Probably an insect]

Mick West

Administrator
Staff member
Given the size in pixels and an assumed size, you can work out
  • the angular size (pixels * FOV /1920)
  • distance (size/2)/(tan (angular size/2)
  • the speed (difference in distance for one frame)*60 feet per second (*3600/5280 for mph)
The FOV seems to be about 40°

This was prompted by this video:
Source: https://www.youtube.com/watch?v=4IvY41SQMZ8&t=1856s

Which is very long and slow, but discusses the math around 30 minutes in. They use an estimated FOV of 38.5.

For something the size of a bird this seems to raise issues, assuming the math and measurements are correct, then given a "bird" height of 0.25 feet (3 inches) the speed comes out at around 500 mph

The smaller things get, the less likely it is that they are travelling so smoothly under their own power, and the more likely it is that it's the drone motion. But to get something to around 20 mph requires a height of 0.01 feet, or 0.12 inches (3 mm), and it's at 4x as wide as it is long, so about 0.48 inches (12mm). This is about the size of a piece of paper from my shredder.
Metabunk 2019-02-22 11-04-55.jpg

Like:
Metabunk 2019-02-22 11-07-29.jpg

This is somewhat consistent with @purpleivan's "slow bug" recreation at https://www.metabunk.org/posts/227878/

I'm not sure how the earlier motion fits into this though.
 
Last edited:

purpleivan

New Member
This is somewhat consistent with @purpleivan's "slow bug" recreation at https://www.metabunk.org/posts/227878/

I'm not sure how the earlier motion fits into this though.
Has someone put together information of the track of the object across the frame, for the duration of its visibility. I could put this together manually frame by frame, but it would be a bit tedious. When I made my video, I just eyeballed the aparent movement in the OP video and set up my animation to approximate it.

I wasn't looking for a one to one match, by just to make the general case of the slow movement across the frame in the distance, with a rapid woosh at the end, being consistent with a bug flying near the drone.

As some extra information, here is an image that shows the movement that takes place in my animation as well as the track my object makes across the frame. Hopefully my image is not too confusing. The longer dotted track is the trajectory of the camera and the shorter, that of the "bug". I added the colour change over time (red to green), of the movement across the frame, to give a better visual.


I increased the size of my object to 300% to make it easier to see in the early frames (the ones in red) at the size the image appears here.

I don't know what else the object in the OP video does in terms of movement, but I'm sure that more complex movement in the early period of its visibility could be replicated in my animation, by just adding greater direction change, and or velocity change, during that period of its movement.
 

deirdre

Moderator
Staff member
assuming the math and measurements are correct,
the measurements are highly suspect. because of blur and the bird at angles.

i'm still confused about "bird height", what part of the brid are you measuring as 'height'. (it's hard to watch that video, i did try to listen but cant figure out where he is getting his 2.2 measurement in that chart. )
h.JPG
 

Mick West

Administrator
Staff member
Has someone put together information of the track of the object across the frame, for the duration of its visibility.
I did this visual track earlier today - it's the large close frame as before, then a "tail" which is the earlier path, encompassing the apparent banked turn. The very start is very indistinct, less than a pixel in size.
Metabunk 2019-02-22 13-16-13.jpg
 

Mick West

Administrator
Staff member
i'm still confused about "bird height", what part of the brid are you measuring as 'height'. (it's hard to watch that video, i did try to listen but cant figure out where he is getting his 2.2 measurement in that chart. )
Just a rough guestimate of what the height of the bird will look like flying towards someone - about from the back to the chest in your picture.
 
The problem here is always the same: we have a 3D scenario that we can but infer from 2D images.

It makes me recall Renaissance's "The Brera Madonna", in which some people claim the egg is hanging above Virgin's head. I think it's easy for anyone here to know the egg must hang behind her head, but this is a geometrical picture, thus much easier than our scenario.


All I can say from Mick's useful frames at #76 is that thing's size in frame 11 is twice its size in frame 8. Thus, its distance from camera at frame 8 must be twice its distance at frame 11.

I used https://sizecalc.com Objects at a distance are measured in angle degrees and by filling viewing distance and physical size with any values, we can see that by doubling viewing distance for same physical size we half perceived size, or by halving viewing distance we double perceived size.

The problem here is that we have no cues for thing's distance at any frame. If we assume it's 1 foot at frame 11, then it's 2 feet away in frame 8.

Interestingly, thing's width is 25% of the whole frame's width (Mick's frame 11), thus 1920/4=480 pixels wide. If we assume thing's distance from camera when it first appears, at 1 pixel size, then for it to become 480 times smaller bigger, distance must be 480 times smaller in frame 11, that is 38.5 feet (if we assume it first appears 3.5 miles away).

I checked the camera is 60fps, as each second is covered by 60 frames. Thing first appears in frame 6475 and Mick's frame 11 is raw video's frame 6569. Thus it takes 94 frames or 1s56'. That would make 8000 miles per hour! to travel 3.5 miles in 1s56'.

Whatever thing traveling at such speed means 195 feet per second and had to be almost that size to appear in each Mick's frames.

These are some lose thoughts for investigation of this flying thing. I don't think 3.5 miles hypothesis likely. How could such big flying thing escape attention of people commanding drone?

 
Last edited:

deirdre

Moderator
Staff member
Just a rough guestimate of what the height of the bird will look like flying towards someone - about from the back to the chest in your picture.
ok. that's what i thought it should be. i couldnt figure out how he was getting 'height' from where you timestamped the video. maybe height he did later when the wings were horizontal.
 

igoddard

Active Member
Again though, shutter speed is not the same as frame rate.
True. So knowing the frame rate provides no clue for estimating object speed?

The study shows the relation of shutter speed (ie, the dark blue Exposure regions) and frame rate (fps spans)...


This then seems to follow from that, that the on-screen distance from the 'right end' of the object appearing in one to the next frame (vertical harsh marks) would represent roughly the distance traveled in 1/60 sec, given fps = 60.

UFO_three-frames.jpg

The changing and oblique angles makes estimations more problematic, but I think not impossible. A rough eyeball estimate is the object travels a distance 2 to 3 times its length per 1/60th second. I would be surprised if this is useless information for testing particular object size and velocity hypotheses.
 

igoddard

Active Member
And frame 10 (with deblur settings shown)
View attachment 36236

Peregrine falcon
View attachment 36237
I'll buy that as the best default hypothesis short of proof. Birds of pray can fly like bullets. Ideally we want to find a known case that matches up. Searching for birds vs drone videos might get an example. There are videos where a hawk will repeatedly attack a drone. So there might be some matching bird zipbys out there.
 
True. So knowing the frame rate provides no clue for estimating object speed?


[...] oblique angles makes estimations more problematic [...]
In Physics motion of projectiles you have a vertical speed and an horizontal speed, which make calculations easier if we use this analogy.

Horizontal speed is what concerns us most here. But we don't know for sure its trajectory to know where to put the referential. Anyway, we can use picture's horizontal line as referential for the sake of simplicity. Once thing makes an unknown angle with this horizontal line, its true speed will be higher than calculated anyway.

Again, problem is always the same: we only have 2D data and this is a 3D scenario. So we can't measure anything on the horizontal referential line because there's no way we can know thing is 1 foot way from camera or 1000 feet...

If thing is flying above tree lines close to camera (red line) its speed will be much slower than if thing is flying above far trees in hill (green line). This makes a huge difference and there's no cue we can tell where thing is flying...

upload_2019-2-23_12-53-49.png
 

Attachments

Mick West

Administrator
Staff member
True. So knowing the frame rate provides no clue for estimating object speed?
Actually, it provides a much better clue that motion blur, as you can measure exactly the difference that a feature moves in a known time. I think that should be clear from the images above.

Using the length of the motion blur (and the shutter speed) is only necessary if you've just got a single image.
 

jarlrmai

Member
if the camera is using the standard 180 rule at 60FPS the exposure time per frame would be 1/120. If it helps this means that in each 16.67ms frame the camera exposes for 8.33ms.

The meta data lists the camera model as the FC550 which is DJI Zenumuse X5 camera which is apparently Micro 4/3rds and can fit multiple wide angle lenses at 12mm and 15mm the included kit lens is 15mm

Here is a link to the product page - https://www.dji.com/uk/zenmuse-x5

So if the object is far away and large the object is fast, really fast, probably hypersonic speeds close to the ground, would have all sorts of other effects (sonic booms) etc it would have been noticed by the people filming, if you've ever had a fighter jet fly low past you it's very noticable.

Now of course all of these things can be discounted by people saying an alien spacecraft can just do those things silently and unnoticed.
 
Last edited:

purpleivan

New Member
I did this visual track earlier today - it's the large close frame as before, then a "tail" which is the earlier path, encompassing the apparent banked turn. The very start is very indistinct, less than a pixel in size.
View attachment 36249
I was actually looking for something that showed the position change relative to the entire video frame, rather than a crop of it, so I could adjust my render to try to match it. To get this I went about extracting frames from the video using ffmpeg, then marked the center of the position of the object on each frame on which I could see it.

The result was this, which shows the position of the object in the uncropped OP video. One difference between mine and yours is that I just couldn't see the object beyond the point indicated as "First frame visible". There's an entire earlier portion of its movement (decending almost vertically in screen space) that just doesn't show up well enough for me to see it, even if I adjust levels to try and make it stand out.


While looking closely at the frames for the first time, I noticed something interesting in the shape of the object around the point at which it changes direction.

I arranged these 5 frames overlaid and offset (frames in time order left to right), to show the change as the object enters the turn. To me it looks if the closer the object got to the turn, the more the extents of it seem to arc back, away from the outside of the turn. So, a lot like the wings of a bird as it goes through a quick change in direction.



I also put together this image that shows all frames of the object, from when it enters the turn, to its last visible frame, and to me the change in the apparent dimensions of it are quite striking.

As the object approaches the turn it is significantly taller than it is wide, and its height is greater than the change in position between frames, which to me suggests that not all of that can be explained by motion blur.

Coming out of the turn, the object swivels round over a few frames, ultimately taking on a distinctly wide than tall
shape. Of course in the last few frames, more of its width can be explained by blur as its horizontal movement in screen space is far greater.



Finally I made this GIF of the direction change section of the video (slowed down to aid clarity). To me this has the broad characteristics I'd want if I was animating a bird during a tight turn.



So based on the above I'm now officially jumping ship and joining Camp Bird :)
 

Ravi

Member
purpleivan, I like the bird theory. However, what you show happens in a fraction of a second. Some birds a quick, but that quick?
 

purpleivan

New Member
Purpleivan, I think you used some tools to get those frames.

This is what I get from simply zooming frame 6543 to about 450% without any extra tools:

View attachment 36265
Not quite sure what you mean about my having "used some tools".

I extracted the frames from the original 4K MKV video using ffmepg to save them as BMP files. After that I found the relevant ones and dropped them into Photoshop, so I could do the work to composite them together for some of the images I made.

On a couple of the images in my previous post, I'd adjusted the levels of the image to give it better contrast, but other than that, these frames are extracted straight from the original video.
 

igoddard

Active Member
Actually, it provides a much better clue that motion blur, as you can measure exactly the difference that a feature moves in a known time. I think that should be clear from the images above.

Using the length of the motion blur (and the shutter speed) is only necessary if you've just got a single image.
That's been my point: "that a known video frame rate could serve to measure the distance an object travels." You've pointed out that blur-span measures shutter speed, not frame rate. That's a vital stipulation I didn't know. But that doesn't mean it's not the case "that a known video frame rate could serve to measure the distance an object travels."

The obvious-bug path could serve as a useful expectation comparator. So I just outlined its frame-rate path...

Bug-path.jpg

On its face, this leads me to believe there's nothing out of the ordinary in the UFO's frame-rate-tracked displacement, given that a mere bug also appears to travel close to the same self-size-referential distance per 1/60th of a second.

This comparator also implies that bug wings get blurred out on close zip-by, causing the bug's shape to become simplified into more of a smooth-pebble shape. This comparator makes me a bit skeptical that the UFO is as large as a bird of prey. I think we should keep open to smaller creatures, like perhaps a dragonfly. Dragonflies can zip around with amazing speed and along very straight paths and make very sharp turns.
 
Not quite sure what you mean about my having "used some tools".

I extracted the frames from the original 4K MKV video using ffmepg to save them as BMP files. After that I found the relevant ones and dropped them into Photoshop, so I could do the work to composite them together for some of the images I made.

On a couple of the images in my previous post, I'd adjusted the levels of the image to give it better contrast, but other than that, these frames are extracted straight from the original video.
Maybe you exaggerated adjustments. You're supposed to use raw images or disclose any adjustments, I think.
 

Whitebeard

Senior Member
purpleivan, I like the bird theory. However, what you show happens in a fraction of a second. Some birds a quick, but that quick?
Falcons are VERY quick, in a stoop (a shallow dive), some can top out at a stunning 200mph
Other species of hawk can manage 130+ mph and even the humble feral pigeon (or stock dove) can do 70-80mph in level flight.
 

purpleivan

New Member
Falcons are VERY quick, in a stoop (a shallow dive), some can top out at a stunning 200mph
Other species of hawk can manage 130+ mph and even the humble feral pigeon (or stock dove) can do 70-80mph in level flight.
I had the same thought about the rapid rate of turn, which makes me thing that if it's a bird, then it might be quite small, as these would be capable of more rapid direction change. However it must be remembered that the apparent degree of change is exagerated due to the angle it is being viewed at from the drone.

In my rendered video the object appears to go almost straight ahead (with some downward movement), then rapidly turns left across the frame, but in fact the object moves in a near straight line, with a slight right turn (relative to the camera) of 15 degrees to avoid there being too much movement across the frame when it wooshes past.
 

purpleivan

New Member
Maybe you exaggerated adjustments. You're supposed to use raw images or disclose any adjustments, I think.
Firstly the extraction of the video frames. This was done using FFMPEG (command line used was ffmpeg -i video/video.mkv images/output%03d.bmp).

The adjustments I made to the extracted frame images were not exaggerated, by which I assume you mean that the adjustment exaggerated some feature of the images, in a way that changed how they might be interpreted. It was a fairly modest levels adjustment, which was added to make the object clearer against the background.

Here are unaltered versions of the ones which received this treatment, shown in my earlier post. Note that the first of these has been enlarged to 200% due to the small size of the object in the relevant frames, but the others are unaltered in this respect. All images below are cropped verson of the original video frames.

Banking object... the arching back of what might be wings is still evident.


An overlay of all frames in which I could make out the object. The was acheived by setting all layers in Photoshop to "Lighter Color" which compares each pixel of a layer with the same in the layer below, then displays the lighter of the two. This was necessary in order to put together this type of composite image.


Here is a single frames used in the composite, so you can compare the result.


Hopefully this satisfies your standards for unaltered imagery/accompanying alteration information.
 
Wings are not as evident here but still seem to be there.

I couldn't reproduce those seemingly wings so far with the ones I zoomed in randomly by trial and error, but there are many such shadows to work. Maybe you could provide frame numbers on your composition "banking object" (or exact time they appear).
 
This then seems to follow from that, that the on-screen distance from the 'right end' of the object appearing in one to the next frame (vertical harsh marks) would represent roughly the distance traveled in 1/60 sec, given fps = 60.

View attachment 36259

The changing and oblique angles makes estimations more problematic, but I think not impossible. A rough eyeball estimate is the object travels a distance 2 to 3 times its length per 1/60th second. I would be surprised if this is useless information for testing particular object size and velocity hypotheses.
I agree that each rightmost extreme of thing marks beginning of a 1s/60 frame.

However, I think there must be a gap between frames, otherwise thing (and bird in #97) would be continuous.

So, each frame might be exposing less than 1s/60, then stop before next frame starts 1s/60 after the one before.

I don't know if I make it clear. I don't have a camera here to test, but maybe if someone can capture a chronograph with precision to 1/100 of second (like iPhone) with a camera of known frame rate (I thing that will be the same irrespective of frame rate) one could check this hypothesis.

Anyway, I don't think this will help a lot debunking flying thing, but at least that will satisfy curiosity.
 

jarlrmai

Member
So, each frame might be exposing less than 1s/60, then stop before next frame starts 1s/60 after the one before.

Yes I have said several times in this thread, that's how digital video works.
 
Thanks for confirming my suspicion, jarlrmai. Sorry to have lost your clarification before. There's so much info here that one must always miss some (sometimes valuable one).

I could make an experiment in the meantime. A digital chronograph isn't very useful here, so I used an analog watch chronograph with divisions of 1/20 (0.05s per division) and a camera rated 24fps (more exactly 23.976fps).

So each frame should follow handle for an entire division. However each frame covers half division at most, then it jumps about a division. One is working with such speedy handles that's it isn't possible to fully quantify frame and jumps, but that's enough to verify your information. I'll reread your 180 rule though.
 

Mick West

Administrator
Staff member
Thanks for confirming my suspicion, jarlrmai. Sorry to have lost your clarification before. There's so much info here that one must always miss some (sometimes valuable one).

I could make an experiment in the meantime. A digital chronograph isn't very useful here, so I used an analog watch chronograph with divisions of 1/20 (0.05s per division) and a camera rated 24fps (more exactly 23.976fps).

So each frame should follow handle for an entire division. However each frame covers half division at most, then it jumps about a division. One is working with such speedy handles that's it isn't possible to fully quantify frame and jumps, but that's enough to verify your information. I'll reread your 180 rule though.
I think you are overthinking this. We know that the frame rate is 60 fps. So we just need to measure the distance it traveled between frames. The amount it travels during the (uncertain duration) exposure portion of a frame is largely irrelevant.
 
That was an exercise unlikely to debunk this in first instance, it was just for the sake of curiosity.

Anyway, that's true each frame marked in #88 bottom is 1/60 sec.

Amount of travel is the hardest part here. We had to know either thing's size or distance from camera and we know none...
 
We can do that by guessing distance.

== Using Mick's frames from #76 and given explanation given in #86 that for a perceived size to be halved, then distance from camera must be double, we know that frame 11 must be twice closer to camera relative to frame 8.

If we assume frame 11 is 1m away from camera, then frame 8 is 2m away from camera. Once that takes 3 frames of 1/60 sec each, then it took thing 1/20 sec, or 0.05s to travel 1m.

That means 5m/sec or 18Km/h (multiplying 5 by 3600).

If that's 100m from camera in frame 11 (and 200m in 8), then speed would be 1.800Km/h!

For any hypothetical distance, we can use this formula to calculate speed (v) from hypothetical distance in frame 11 (x):

upload_2019-2-25_0-18-22.png

== Then we have size. Thing in frame 11 covers 25% width of frame, that is 480 pixels. We know from camera specs given by jarlrmai in #92 (DJI Zenmuse X5) that diagonal FOV is 72º. Then, perceived size is 18º (25% of 72º). Filling in data for viewing distance and perceived size in www.sizecalc.com we have 0.32m as physical size for thing (that is about a third of [any] hypothetical distance for thing in frame 11) :


upload_2019-2-24_22-34-5.png

== If we take into account that drone operators didn't see anything abnormal at the time, we can guess thing couldn't hardly be much larger than, say, some 40cm (16 inch). Maybe something as large as an eagle, as they might have seen 2 and don't recall that, I think (or don't associate to this thing). Anyway, that makes it unlikely that thing is 100m away from camera in frame 11, as it would not only be traveling 1.800Km/h, but would also be some 30m long (very hard to miss).

So far, all we have are hypothesis that we can't test. But we can think how likely is each one...

[Sorry for SI units, but I'm not very used to US units]
 
Last edited:

deirdre

Moderator
Staff member
If you assume a given size then the distance, and hence speed, can be calculated.
an American Kestral was spotted in beaver County in December
They are tiny little falcons. Can you determine distance if you know their size? from that frame formula? He'd obviously have to be alot closer to the camera to start, or does that not matter in that formula?

(a tiny bird would also explain how in the very first frames it looks liek he gets blown back (or sideways) a frame)
kestral.JPG


The only speed i can find is horizontal flight. which is up to 39mph compared to the pergrine's 55 mph.
 

purpleivan

New Member
I'm curious where the figure of 60fps is coming from for the video.

The two sources I have for the framerate (the output of ffmpeg, when extracting frames to images) and MKVToolNix comes up with a framerate of either 24 or 48fps.

When I first extracted individual frames from the video, I by mistake extracted it explicitly at 30fps. This resulted in a duplicate every 5th frame, but when extracting without a framerate set, there were no duplicates. This suggested to me that it was a 24fps fideo.

Looking at the video in MKVToolNix, the relative frame timecodes alternate between 41 and 42 milliseconds per frame for track 1, indicitive of 24fps.

However there are also timecodes for a track 2 shown in the data MKVToolNix provides, the frame times for which lie in the center of the times for track 1. Still 41/42 milliseconds apart, but offset by 21 millis.



I have to admit I'm a bit out of my depth regarding knowledge of MKV video format structure, but these frame times suggest to me a framerate of either 24 or 48fps, depending on the meaning of the track 2 information.

Is the 60fps figure mentioned earlier based on the stats for the camera used to record the original video. If so the discepency could be due to a the framerate being changed when the original from the drone was processed to have the subtitles added.

Regarding the queston that Kasparovitch had about the frame numbers/elapsed times of the "wing like" curvature I could see in some of the frames, the only times I can give are based on the 24fps that ffmpeg output the video into image files at.

The frames (at 24fps) composited in the image below were 4672 - 4677 inclusive.


I've also put a set of the frame images extracted using ffmpeg, where I could make out the object on Dropbox here.
 

Mick West

Administrator
Staff member
I'm curious where the figure of 60fps is coming from for the video.

The two sources I have for the framerate (the output of ffmpeg, when extracting frames to images) and MKVToolNix comes up with a framerate of either 24 or 48fps.
Where did you get a .MKV file? The "official" raw video seems to be this:
Source: https://drive.google.com/file/d/1QVGWQI46qbt4UkK1_keYne4-NXUwrLQe/view


Which links to a source file DJI_0004.MOV, which downloaded as
"RAW FOOTAGE of UAP Sighting_ Beaver, Utah (October 18, 2016).MOV.MOV"

Which is a 1080P 59.94 FPS H.264 file
 

purpleivan

New Member
Where did you get a .MKV file? The "official" raw video seems to be this:
Source: https://drive.google.com/file/d/1QVGWQI46qbt4UkK1_keYne4-NXUwrLQe/view
Which links to a source file DJI_0004.MOV, which downloaded as
"RAW FOOTAGE of UAP Sighting_ Beaver, Utah (October 18, 2016).MOV.MOV"

Which is a 1080P 59.94 FPS H.264 file
Argh... completely missed the link on the YT page, I've been using the YT video itself, the 2160p version. However it would seem that all that additional resolution contains is some nice sharpness to the lettering in the subtitles.

That explains a lot, mainly the difference in framerate which as I suspected would have been introduced when the text for YT was added.

I'll take a look at the video on Google drive.
 

purpleivan

New Member
So now that we're all in sync (more to the point, I'm in sync with everyone else), here's an updated image, showing what I think shows the turn in flight of a bird.


I still think that the arc of the wings mid turn is fairly pronounced, especially in the range of frames 6545 - 6552. But the main point would be the clear change from a tall thin, to short wide aspect ratio, just as the object changes direction, which to me suggests an object with clearly identifiable wings, that are not beating during the turn.

So most likely a bird rather than a bug (contrary to my original opinion on this), as insects would not exhibit such a distinct change in ratio, due to the high movement speed of the wings.
 

Gerard

Member
So now that we're all in sync (more to the point, I'm in sync with everyone else), here's an updated image, showing what I think shows the turn in flight of a bird.
I too thought that the video, especially the banking turn frames you reference, looked more like a bird than a bug.

However the object distance/size/speed information that's been discussed in a number of posts here appears to rule out a bird-sized object while being compatible with a bug-sized object.
 

deirdre

Moderator
Staff member
So most likely a bird rather than a bug (contrary to my original opinion on this), as insects would not exhibit such a distinct change in ratio, due to the high movement speed of the wings.
i noticed too that right as we see the "bird" appear, (or when i can see it) it moves to our right but using the frame by frame, it moves quickly back to the left for one frame. which to me is saying it must be really light weight. and i don't think any birds with those ratios would be light enough to get blown sideways so quickly. I'm leaning toward piece of fluff.
 
If you assume a given size then the distance, and hence speed, can be calculated.
We can do that by guessing distance. [...]
After discussing size of thing, we can now discuss likely speed.

Thing first appears on frame 6475 and last on frame 6569, so it takes 94 frames to travel distance. 94 frames at 1/60 means 1.56 secs.

If we assume maximum distance for thing is 3.5 miles away, then it would travel 3.5 miles in 1.56 sec, or 8.000 mph, not really likely.

If we assume maximum speed to be 100mph, then it travels 0.028 miles (=147 feet) in a second, or 230 feet in 1.56 sec.

Then 230 feet is perhaps the farthest possible distance from camera.

Given thing is 480 times nearer to camera in last frame (6569) as discussed before, that would be 0.21 feet (or 2.5 inch) away from camera.

Using www.sizecalc.com assuming viewing size is 18º on last frame (6569), as thing covers about 25% of frame width here and camera angle is 72º, and distance from camera is 2.5 inch, then physical size would be 0.8 inch!

upload_2019-3-8_21-41-54.png

Now we must allow for image spreading during framerate and we perhaps should divide that size by 2 or 3, which makes thing 0.4 inch at most, if it travels at 100mph.

If it travels at slower speed (as it likely does), then its size will be proportionally smaller.

So I think thing must be a bug. A white bug...
 

CeruleanBlu

Active Member
With the last decade in advancements of both commercial and hobbyist drone, fixed-wing and multi-rotor technologies, I'm still a firm believer in finding more mundane explanations for unusual things.

From relatively simple construction, yet crammed with tech agricultural multi-rotors such as this:


To RC airplanes capable of over 500 miles an hour, such as this:


There's simply no end to what mankind can create for fun to throw the UFO believer crowd into another quest for the truth. And until they are ready to accept the simple man made things we're capable of producing, there may be no real understanding.
 

Mendel

Member
[...]
I really hope someone can get to the bottom of this. Here is the Metadata. The time, date and location match to the uploaders claims:

File Name RAW FOOTAGE of UAP Sighting_ Beaver, Utah (October 18, 2016).MOV File Size 1235 MB File Type MOV File Type Extension mov Mime Type video/quicktime Major Brand Apple QuickTime (.MOV/QT) Minor Version 2014.2.0 COMPATIBLEBRANDS 0qt Movie Data Size 1295120164 Movie Data Offset 40 Movie Header Version 0 Create Date 2016:10:18 11:19:22 Modify Date 2016:10:18 11:19:22 Time Scale 60000 Duration 0:02:52 Preferred Rate 1 Preferred Volume 100.00% Preview Time 0 s Preview Duration 0 s Poster Time 0 s Selection Time 0 s Selection Duration 0 s Current Time 0 s Next Track Id 2 Gps Coordinates-Err 38 deg 15' 55.59" N, 112 deg 36' 44.11" W, 3.1 m Above Sea Level Speed X-Err +0.70 Speed Y-Err -0.20 Speed Z-Err +0.10 Pitch-Err +0.50 Yaw-Err +170.50 Roll-Err +2.50 Camera Pitch-Err +0.00 Camera Yaw-Err +171.90 Camera Roll-Err +0.00 Comment 0.9.198 Category v01.27.5134 Model FC550 Track Header Version 0 Track Create Date 2016:10:18 11:19:22 Track Modify Date 2016:10:18 11:19:22 Track Id 1 Track Duration 0:02:52 Track Layer 0 Track Volume 0.00% Matrix Structure 1 0 0 0 1 0 0 0 1 Image Width 1920 Image Height 1080 Media Header Version 0 Media Create Date 2016:10:18 11:19:22 Media Modify Date 2016:10:18 11:19:22 Media Time Scale 60000 Media Duration 0:02:52 Handler Class Media Handler Handler Type Video Track Handler Description DJI.AVC Graphics Mode srcCopy Op Color 0 0 0 Compressor Id avc1 Source Image Width 1920 Source Image Height 1080 X Resolution 72 Y Resolution 72 Compressor Name Dji AVC encoder Bit Depth 24 Video Frame Rate 59.94 Gps Coordinates 38 deg 15' 55.59" N, 112 deg 36' 44.11" W, 3.1 m Above Sea Level Speed X +0.70 Speed Y -0.20 Speed Z +0.10 Pitch +0.50 Yaw +170.50 Roll +2.50 Camera Pitch +0.00 Camera Yaw +171.90 Camera Roll +0.00 Avg Bitrate 59.9 Mbps Gps Altitude 3.1 m Gps Altitude Ref Above Sea Level Gps Latitude 38 deg 15' 55.59" N Gps Longitude 112 deg 36' 44.11" W Image Size 1920x1080 Megapixels 2.1 Rotation 0 Gps Position 38 deg 15' 55.59" N, 112 deg 36' 44.11" W Mime Type video/quicktime Category video Type quicktime Raw Header 00 00 00 20 66 74 79 70 71 74 20 20 20 14 02 00 71 74 20 20 00 00 00 00 00 00 00 00 00 00 00 00 4D 31 F7 2C 6D 64 61 74 00 00 00 02 09 10 00 00 00 20 06 00 0D 80 B8 92 FF F7 36 00 B8 92 FF F7
  1. Do we have this metadata for each frame? If so, What was the drone's flight path, how fast was it moving? This is especially important as some posters are considering an object close to the camera, which would be moving relative to drone speed.
  2. What was the sun's position at the time? With respect to the heading of the camera? The angle that the object is lit up could be helpful.
  3. Is it still possible to get satellite cloud data for that time for that area? This would be helpful, because the object is clearly visible in sunlight and might be invisible against the dark background if unlit (or should at least show a change in brightness), so knowing which part of the terrain was in shadow would indicate where the object could NOT have been at the time. (It might also help locate the dark object entering the frame from above a few seconds earlier.)
 
Top