Photo-Surprise

reply to post by internos


Are you mad because I didn't like your picture or because I came up with more data then you?

I do appreciate your perspective though, who knows it could be a bird. But the problem her is the attitude, if we don't automatically say "oh it could be a bird or a bug" right away, people assume we are "ignorant".

Speaking of "trash", so am I to get this right that you are likening me to a "troll" because I disagree with you and dispute your data? I just want to make sure we're clear on that. Or is because I illustrated how we(those of us on the board) hear the same analysis blanketed over every situation first as fact?

Either way man, if anyone has proven themselves to be the "IGNORANT", "ARROGANT", and "WRONG" in the way the present themselves and their ideas/data it's you. And if you are acting "SANE" and "open-minded" I'm glad I'm crazy and focused.

If you want to "waste more of your time", by all means Or if we're going to discuss who's more of a "poopy-head" we should take it somewhere else.


[edit on 8-3-2008 by Shakesbeer]

There are some big flying insects in the south of France, including the green scarab cetonia aurata and cicada.
The green scarab has a higly reflective carapace - looks metallic. It is quite heavy for an insect, and moves its wings very fast. This could explain the darker shadow. The upper part of the carapace moves wide open to uncover the wings when flying, which could explain the odd shape.

(edit : removed non-working picture link)

I am not convinced by the bird or insect explanation... too weird.

[edit on 2008-3-8 by nablator]

reply to post by nablator


Link doesn't work or is my computer being lame?

reply to post by Shakesbeer


Don't take the words as addressed to you personally because they were not, i'm sorry if you think so, but it doesn't matter: re - read the post, and you will realize that i was repling to your post, AS IT WAS DESERVING, nothing more. If you express your opinion, then it's ok, who cares if you disagree with me? I often disagree with my girfriend/s. The problem is that you addressed your posts not only against my opinion, but even against me, PERSONALLY. This is why i've been rude. How does it work: when you attack someone, don't you expect a counterattack?
Anyway, PEACE, let's back on topic, now...

damn those birds and/or insects...we should just exterminate the whole lot of them so when something does turn up on our photographs we'll know for sure it was an interplanetary interloper. No, wait it could be leaves blowing in the wind too; or trash...Also, ground all manmade flying objects...


ahhh...

just kidding, I love critters, mostly; also, riding on planes, except I feel like kissing the ground when its all over.

Photoshop.









DUN DUND DUNDUNDUNDUNDUNDUNDUNDUDNUDUDN I believe its photoshopped.

Hi Guys.

I have just run off some shots for general comparison purposes here for you.

All shots are taken using a Canon EOS 40D using a 70-200 F4 L IS USM lens in full manual using raw image format.
Each has obviously been converted to jpg so you can actually see them using whatever browser you favour using.
Jpg compression is minimum and all crops are 100% unless otherwise anotated.

Exif for all are more or less the same and Ive used an obscenely high shutter speed (1/3200) so motion blur is not an issue.

Everything you see is a result of 'misappropriated' manual focusing shot through a really mucky window...

I have also separated out the color channel information in a few of them for you, so you can compare and contrast.

Follows the Exif:


Canon EOS 40D
F: 6.3
ISO: 400
FL: 180mm-200mm
Ex: 1/3200
Manual Exposure
0 Exposure Comp.
Image Stabilisation is on
Color Space: sRGB (24bit) Converted from CR2 Raw.

Birdie 1: In Focus (More or less)


Birdie 1: Out of Focus


Birdie 1: Very OOF.


Birdie 3 OOF


Birdie 4 OOF


Birdie 6, OOF and a 50% Crop.

Now...
Color info for your perusal.

Color shot OOF Object.


Same Object Blue Channel.


Same Object Green Channel.


Same Object Red Channel.


Object 5 Seconds earlier when it was in focus and wasnt in a cloud...

Color:


Blue:


Green:


Red:


Note that in all cases the color channels are more or less even and all objects whether in focus or not display a distinct absence of 'funny looking triangular shadows'.

Feel free to pick them apart at your leisure.

Absence.

reply to post by Absence of Self


Hi there,

First, I want to thank you for you time spent doing this.

I have a few questions...

-Coud you give me the detail of the "analysis" you did (calculus) ?
(In order for me to have it peer rewieved & verified)

-How come you do not take into account the angular velocity or did you ?

That's it for now.

Now, for the ones who tried different color channels, dumb question, what is the name of this kind of test and what can you learn from it ?



Peace,
Europa



EDIT : English is not my 1st language but I am trying my best


[edit on 9-3-2008 by Europa733]

reply to post by Absence of Self


Hi again,

would you mind doing the same experiment with : en.wikipedia.org...

Just curious...

THX.
Europa aka Buckwild


EDIT ; If anyone know an image analysis laboratory who would be interested
in studying this case, please PM me as I think this picture holds a lot more
data that one could imagine.


[edit on 9-3-2008 by Europa733]

Hi Europa,
Yeah, no problem ~ possibly...
What I was effectively concentrating on was a combination of 3 factors associated with the image.

1. Possible resultant image 'depth of field' for the camera at the settings recorded by the image exif (allowing us to gauge what is in focus and the distance limits involved).

2. Field of view for the lens at the time. (effectively giving us a triangular measurement allowing us to calculate "possible" image widths at various distances; the minimum 'in focus' distance being determined by the DOF)

3. The optical characteristics of the image itself as recorded by the sensor. (i.e. its channel data ~ though it is a given that it has been reduced by the post processing chip within the camera when the information recorded by the sensor is translated from the CCD bayer array to the output image format by the jpg encoding algorithm)

(velocity measurements are tricky so i'll cover that one later)

1.

Image DOF is a handy calculation normally based upon a combination of your lens optics, f-stop and sensor factored by the distance of the focal plan of the lens from the sensor and the design of each, which effectively gives you a good idea of what that lens/sensor combination can accurately see (resolve) for a given focal length and object depth.

Good explanations and tutorials are available for you here:

Cambridge In Focus
and here
Luminous Landscape

If you want to go crazy and run the arithmetic I would suggest you start here with Norman Koren's tutorial:
Norman's DOF Intro

(though you may not want to, the maths can be kind of confusing)

Sorry, I'm outta time.

I'll do more stuff later with regards to the FOV calculations and image information but for now I'm due in the pub.

reply to post by internos
Hi internos,

I have the utmost respect for your work in this forum, your knowledge of computers and doing the things you can do are to me at least, very impressive. Please don't think you are wasting your, or anyone else's time here, your selfless work is most appreciated by me and countless others.

I don't claim to have super vision, in fact I wear reading glasses, however, aside from reading glasses, at the time of, my last vision check, I have 20-15 vision. In the original picture at least to me, the object does appear several hundred feet beyond the clock tower, but in front of the hill/mountain in the back ground. right or wrong in that, I had no intention of insulting you or anyone else in this thread, my opinion was merely that, my opinion. I do not believe it's a bird or an insect, I just don't see the lines you have outlined in your post in this thread. However, keep up the great work you do, for those of us that don't have the knowledge or talent you have.

If it is fake, the artist really understood not only light and shadow, atmospheric simulation, perspective manipulation, and would have to have not been so over zealous as to make the "craft" look too apparent. Especially considering you get a true sense of volume around the object, that you can obviously tell it's moving through the atmosphere.

Now anyone who has actually worked with digital imaging and 3D animation can tell you there isn't a "Make a flying saucer" button you can push let alone make a convincing one. Someone would have had to spent a lot of time on this image to fake it.

One thing that does still strike me as very peculiar though is the shadow. Why is it showing up as a triangle? The peak seems to stick out above the object which would not coincide with the angle of the sun, be it a bird of a probe-lovin' alien At the same time I see absolutely no digital signature on the original jpeg that would lead me to believe any of this was added later.

Originally posted by Shakesbeer
One thing that does still strike me as very peculiar though is the shadow. Why is it showing up as a triangle? The peak seems to stick out above the object which would not coincide with the angle of the sun, be it a bird of a probe-lovin' alien

At the same time I see absolutely no digital signature on the original jpeg that would lead me to believe any of this was added later.

Hi there,



One of my "theories" is that X could be a triangle spinning on it's axis
at a very fast rate.

Just my two cents...

Peace,
Europa

Definitely Not a bird.

Too big, too blurry, and too reflective and light blue in color, I would say this is the least likely explanation.

Insect maybe. Rapid moving wings could explain the shadow. An insect zipping past a few feet away, with the camera focused in the distance makes sense to me. The insect would be very blurry, an elongated as well, and the exoskeleton does often come off as metallic.

Over a decade ago now, a news crew was filming a report in San Francisco with a shot of the city in the distance, and an insect flew past behind the reporter, and it looked like a UFO zipping past over the city. The film got quite a bit of air time over a short period, until someone came up with the reasonable explanation, which through greater examination was well accepted.

What makes me think this is not some sort of aircraft is the angle or attitude at which it is flying. It does not make any sense that a fast moving aircraft would be in such a tilted attitude flying at a high speed over a residential area. It seems too steep of a maneuver for any moving aircraft at what would be a low altitude for an aircraft. It seems that it would have to be moving very fast to be this blurry. It would either have to be landing or taking off, and I find it extremely hard to believe that there is a base for this craft in the area.

So I vote insect.

One possibility is that reflective stealth technology could be malfunctioning, revealing the object in a distorted manner, and the attitude we see is not the true attitude of the craft.

reply to post by Europa733


hmm interesting theory. Let me see if I can do any visualization with the image based on that theory.

Hi Guys... OK back. More or less pissed now but still my reasoning continuing from last time.

2. Lens FOV.

The field of view of the lens in every case that uses a camera is vital to understanding whatever it is you are looking at.

In this case im using the generic formula since i have no real focal multiplier information for that camera and im assuming that it really is a 35mm equiv (OK... im not really. for the calcualtions ive given i've assumed its a 1.6 crop i.e. the generic APS-C format that appears to have taken over the 35mm's).

So... FOV = 2 * arctan (frame size/(focal length * 2)) where framesize is designed by the manufacturer to produce the same as a 22.2 x 14.8mm 1.6 crop multiplier sensor and FL is 7.4 mm from the exif.

Yup, could be complete bull but this is just guesswork and im guestimating the angular FOV according to the majical marketing term 35mm equalavent.

Anyway.

FOV...

Our eyepiece forms what is effectively the sharp end of an isosceles triangle and if we know the angle of separation between the sides nearest the apex (your eye point or the vertex angle) we can determine the picture width for any given distance (if its lower than 180 degrees obviously since that = infinite).

c squared (image width for the distance hypothesised) = a squared + b squared where a and B are equal given the fact that our FOV is an isosceles triangle (and so shall in this case be R), the trick here being that we dont know R: all we know is the 'height' of the triangle since we have already assumed that as a deduction from the minimum DOF given in point 1.

The maths being...

h / (1/2 cos theta) = R: where theta is the vertex angle and h is the distance to the subject. (R in this case is one of the two long sides of the triangle)

and

x = R sin (theta /2) where theta is the vertex angle and x is one half of the 'width' of your triangle and R is the length of the side you've just calculated.

So... 2X is the total width of the image at a given distance R.

Cool..

Now all we are left with is effectively pixel counting.

If we know the image is 1000 pixels wide and this angle covers a distance of (say for example) 1000cm at a distance of 2 meters we can easily deduce an object of 10 pixels width to be 10cm wide at the same distance...

Easy wot?

Now... real voodoo.

Velocities.

**Note** this involves more estimations than you may be comfortable with.

The velocity of an object within an image depends entirely on the FOV and estimated distance as stated above in combination with the exposure time and contrasted blur trace.

Say for a rather extreme example you have an image trace of 100 pixels covering a total image frame of 1000 pixels, 10 of which were accumulated at one end of the trace and of higher contrast / saturation than the others, with the image being taken within a shutter speed of 1/100 of a second then you can safely guess that the object traversed that area of the camera sensor within that time. And.. If you have already calculated the area that particular image artifact covers being 10 cm (1cm per pixel) then you have an area time basis and the velocity is distance covered by object/exposure time.
So 10cm object covering 100 cm in 1/100 seconds = 10cm object covering 10,000cm in 1 second.

But..
Yes. Too much guesswork.
Which is why i dont normally do it and in this case the shutter is 1/400 leaving no motion trace worth considering so i didnt bother considering it.

Color stuff next..

Originally posted by Shakesbeer
reply to post by Europa733

 

hmm interesting theory. Let me see if I can do any visualization with the image based on that theory.

Hi there,

Here is the 2nd "theory" :



X is "seen" twice due to motion blur.


To absence of self : Thank you very much for everything


Peace,
Europa



[edit on 9-3-2008 by Europa733]

reply to post by Europa733


This is not a video, you don't get to see two interleaved frames in a single shot IMO. There's a structure there that's not easily recognizable.

What's bothering me is the predominant blue channel. Why is it so blue? Is it a blue bird? Or a huge mothership very far away? Very puzzling.

No worries Buck.

The color information stuff was more of a side line, albeit an unsual one.
You can see from my images above that most objects (whether in focus or not) exhibit a trace along all color channels within the image (Even the aeroplane and that when OOF is filtered through bad glass about 1.5km of haze and a cloud..).
Mostly because the luminance is even across all channels.

The exception being the image in question.
Here we have more or less a single color channel with an object exhibiting most of its properties within that channel.
(there is a little green but not much)

It appears to be almost absent in terms of its red and green information, which to say the least is somewhat odd. Legitimate reasons for which, i'll admit that i dont have any.

That is not to say that red and green should not be present. They should be, in exactly the same manner that the sky is not just blue.... Its is afer all the sky we're talking about and not a painting.
(Massive compression and full channel dumping.. dunno. It would make sense for a 48 bit image with Alpha i suppose. Not a natural jpg, maybe a tiff. Never tried it so i cannot tell, Maybe JRitzman.. ask him when he comes back..)

On a side note though...

Fake land..

If i was going to fake something i would probably take a decent quality background image and then add a luminance map to it somewhere.. resulting in the color displacement characteristics you can see.

However, if i was going to fake something i wouldnt have included the shadow...

P.s.
The bug you have in question is 2 cm long...
Assuming a .25 displacement for the focal error (i.e. its 125% of its normal size due to fuzziness etc).The image would be about right for an object of 2.5 cm at a distance of 2 feet approx... Its a guess though, i havent done the maths to pin it down but as you can see from my above posts there is a lot of assumption here..

*edit.. though it really cannot be that bug not unless the bug in question is an anorexic.. its too thin.
Bredth proportion is good, height proportion is bad.

There really is not a lot more i can say on the matter.

Cheers
Absence.


[edit on 9-3-2008 by Absence of Self]