Strange streak discussion: 2004 Dec 7 APOD

[J_Joy]

YI, I've send a list of questions to Wayne - which he indicated he would answer - however, I received an automated reply stating he would be out of the office until after the new year. I'll follow up with him then.

JJ

[Guest]

You mean because the last picture is a fake?

Right.
But four minutes to post -
...you must've seen it before.


HAPPY HOLIDAYS, everyone.
Here is my Crisstmmas card to you.

Feel free to write your own message in the stars for all your email friends.



I've sent it to you by the fastest service known:

[HawaiiArmo]

Doc Bluto, now that we understand each other's point, and obviously will probably have to agree to disagree, I'd only like to point out that you hold an emotional sway to the hoax theory. As I said, there is absolutely no way to prove you or me to be correct outside of Sodium Pentathol and a lie detection machine, but you still assume the hoax is more probable. From your point of view, you think 3 + 3 = 6, while the non-hoax point of view 3 + 3 = 7, but I ask you, now who's being emotional about it. There is absolutely no way to assert that.
Besides, as was stated a few posts before, we already had this discussion on a hoax early in the message board. Most of us posting theories started from the assumption that this was not a hoax. We full and well accepted the possibility but dismissed it. Just as you admit it's a distant possibility for this not to be a hoax, we also assume the same. Therefore, there is no point to further argue among these lines of reasoning. If the photographer came to the message board, and said, "I'm just kidding guys, it was a hoax, this is how I created it", nobody would argue otherwise. I'd be willing to give you credit for your foresight and adroit thinking, then again, there is no clear way for your assertion to change. If the photographer said, "it's not a hoax I swear it on my mother's life", that's still not proof of his word, and you definately would not accept it. In other words, for those who believe it to be a hoax, there is no way to prove it otherwise. When the platypus was discovered and it's pelt taken back to the UK, taxonomists laughed, thinking, there is no way an animal like this exists. THey figured it was a hoax, but all it took was a live speciment to disprove the hoax. Unlike that example, we don't have that luxury. So we must start at a theoretical assumption, with no way to disprove the hoax (and thus include it as a possibility, but perhaps not a likely one).

[Can't use my Bad Buoy]

Doc Bluto....snip, bite, and claw......

Christmas Eve day.

What pathetic life could be so consumed by this petty bickering,
now part of the recorded world history.

Our Borg Collective is alive and functional.

You, and some not so young of your ilk, need to learn better restraint and focus.

[Guest]

This Trolling business is disturbing, I must say. http://www.urban75.com/Mag/troll.html

Oh, Please Stop

My sides are aching

I didn't know TROLLS could be so entertaining.

Hmmm, a number of those traits myself.....I didn't realize it was such a career.

[wombat]

Back after two weeks holiday.

I wish to take issue with the someone (about 50 pages back) who made the very negative remarks about "amateurs", then posted some even more negative synonyms allegedly from a dictionary.

Well, as anyone who studied Latin, and the tens of millions of people who still speak Italian now will know, the word "amateur" is derived from the Latin "amo", meaning "to love". It is a direct anglicisation of the Latin "amator" meaning "lover".

Thus, an amateur is someone who does something for the love of it, as distinct from being paid for it. There is no intrinsic negativity in the definition at all. I am amazed that this supposed dictionary had these alleged synonyms in it at all. In fact I don't believe it at all.

It follows, then, that every single one of us who is involved in this debate, even those who are professionals in a related field, is doing so at an "amateur" level, because none of us are being paid or otherwise materially rewarded for doing so. Unless you wish to start an even grander conspiracy theory...

[Cloudbait]

I wish to take issue with the someone (about 50 pages back) who made the very negative remarks about "amateurs", then posted some even more negative synonyms allegedly from a dictionary.

Well, as anyone who studied Latin, and the tens of millions of people who still speak Italian now will know, the word "amateur" is derived from the Latin "amo", meaning "to love". It is a direct anglicisation of the Latin "amator" meaning "lover".

Thus, an amateur is someone who does something for the love of it, as distinct from being paid for it. There is no intrinsic negativity in the definition at all. I am amazed that this supposed dictionary had these alleged synonyms in it at all. In fact I don't believe it at all.

As anyone interested in etymology knows, you can't rely on a word's origin to define it. "Amateur" does indeed mean someone who pursues a subject non-professionally (presumably for love, as the word root suggests). But any good dictionary will also give another definition of the word as a person lacking skill or ability in a discipline, and that definition is as solidly grounded in usage as the first.

It appears to me that amateurs fitting both of these definitions have been involved in this discussion, but only the second type have used the term disparagingly.

[wombat]

Perhaps those more familiar with how these three interact can answer this question.

I'm wondering if we can assume that the central part of the flashed insect image, the part that looks yellow, really is yellow. Even though someone hundreds of pages back showed that the flashed insect image is not saturated, can we assume that the response of the CCD or CMOS (whatever) chip is linear enough that it will faithfully capture the RGB components in the right ratio under all the lighting conditions in this image?

For example, is possible that the central part really is red, or brown, or orange, (or many other colours, for that matter), but that some aspect of the electronics involved has changed it?

I'm thinking of flash photos of things like interiors of houses that you see, where there is something of a single colour (eg a wall), but parts of it are close to the camera and flash, whilst other parts are further away. As you go along the wall in the image, there is not only the expected change of brightness, but also a significant change of colour. From what I can remember, typically the further (and therefore fainter) part will look a lot greener. This is very obvious in emulsion type images, but it also happens in digital images as well.

Finally, since most insects that I have seen are to some extent "shiny", I would expect that there would also be a specular reflection of the bluish-white flash that will "dilute" the true colour by some amount.


So to rephrase the question, how confident are we that this really is a yellow insect????
PS. Cloudbait: thanx for your comments re amateurism. That last line about the two sorts of amateurs was particularly clever. LOL.

[HawaiiArmo]

Doc Bluto....snip, bite, and claw......

Christmas Eve day.

What pathetic life could be so consumed by this petty bickering,
now part of the recorded world history.

Our Borg Collective is alive and functional.

You, and some not so young of your ilk, need to learn better restraint and focus.

Maybe it's not Christmas Eve day for me (In other words, take into account the international dateline, and that some people are still a day behind). Another assumption that's faulty at best. Thus supporting what I was initially saying, everything's relative to each person (a hoax to someone is not necessarily a hoax to another).

[Guest]

So to rephrase the question, how confident are we that this really is a yellow insect????

Though it looks harsh and bluish to you, flash is designed to as closely approximate solar illumination as possible.
So if illuminated solely by the flash - the bug is yellow.

[Guest]

You mean because the last picture is a fake?

Right.
But four minutes to post -
...you must've seen it before.


HAPPY HOLIDAYS, everyone.
Here is my Christmas card to you.

Feel free to write your own message in the stars for all your email friends.



I've sent it to you by the fastest service known:

Thanks, it's a great card. But Optus is an Australian company. What's with the snow and elk?

Anyway, my friends love it, as I did.

[Bob Peterson]

What is APOD up to now? I think what we have here is a Cosmic Turkey. Note the similar features to our discussion photo: a streak, loops, a flash, and smoke-like stuff. Eerily similar and the image was captured by the orbiting Chandra Observatory(sort of like the Canon PowerShot). Take a look.

http://antwrp.gsfc.nasa.gov/apod/ap041223.html

Wish you Happy Holidays whenever they may occur.

[Ernst Lippe]

We have all seen that fakes be generated in a few minutes which are just as
good as the bug simulations carried out with much more time. Clearly the
possibilty that it is a fake it has to be accepted by any scientific minded
person.

Now let's see how likely this possibility is.

First of all, Wayne Pryde has been very low key about this. As far as we
know he has never claimed that it was a meteorite and he never claimed that
he saw anything. This is not really expected behaviour for a faker, because
their goal is to make the event more sensational.

This photo has been surrounded with a fair amount of publicity. If
the photographer had a reputation of being a prankster, this would
probably have been discovered by now.

It is indeed possible to duplicate the effects picture with some image
manipulation program. But it is not at all easy to hide this fact.
JPEG is a lossy compression format. When you edit a JPEG file you first
have do decompress it, then edit it and finally compress it again. In
the compression step you loose information and this introduces artefacts
in the file (e.g. in general it will loose sharp details). Lots of people
have been studying these files but no one has found any evidence that
these images have been digitally altered. Many image processing programs
remove or alter EXIF information that the camera has added to the image
or insert their own information. Again there is no evidence whatsoever
that this happened. So when this image was indeed digitally altered
it must have been done by a fairly skilled person.

But when someone would try to create a hoax image of a meteorite it seems very
unlikely that they would come up with an image like this one. The dark trail
does not extend to the boundary of the image, it starts somewhere in the
corner. But when it was a meteorite the trail must have crossed the boundary
and it would have been trivial for a faker to add this part. Why is the trail
darker, "everybody" knows that meteors leave a light trail. The explosion
does not line up with the light pole. The shape of the explosion is extremely
peculiar. If you ask a random person to draw an explosion it seems highly
unlikely that they would draw anything like this, in particular the strange
shapes above and below (the "wings"). Why did he choose a site where there
would be observers in the neighbourhood and that could be inspected? There are
lots of very quit places around Darwin where he could make a picture without
these problems. So, when it is a hoax it is a very strange one, it would
have been trivial for a hoaxster to create a much more convincing picture.

So the fake theory seems pretty unlikely.

[Cloudbait]

I'm wondering if we can assume that the central part of the flashed insect image, the part that looks yellow, really is yellow. Even though someone hundreds of pages back showed that the flashed insect image is not saturated, can we assume that the response of the CCD or CMOS (whatever) chip is linear enough that it will faithfully capture the RGB components in the right ratio under all the lighting conditions in this image?

For example, is possible that the central part really is red, or brown, or orange, (or many other colours, for that matter), but that some aspect of the electronics involved has changed it?...

So to rephrase the question, how confident are we that this really is a yellow insect????

The color could easily be very different. Reconstructing accurate color is a tricky business, and this image was made under difficult conditions. The color temperature of a flash is similar (but not identical) to that of sunlight. But the primary source of light in this image was from sunset light, which is much cooler. If you look at the white balance settings for digital cameras, they usually cover the range of bright daylight, clouds, tungsten, fluorescent, etc. This image was made using white balance 0 (whatever that happens to mean), and since the color of the sky and harbor look about right, it's a good bet that the white balance was quite far from optimal for the flash. The actual color of the suspected insect could range from light green, through yellows, light orange or brown. It would be a mistake to try and identify (or disqualify) a particular species based on color alone. It is possible to back out the color balance corrections used and apply values for the flash instead. This would yield pretty accurate color, since the CCD itself is very linear and none of the channels are saturated. Doing this would require a more detailed understanding of the G3 than I have.

[MrMoon]

Ok, in that terms it seem unlikely. but...

Taking a picture and being lucky cathcing a bug in such a way that it produces that trail wich happens to end just in the only malfunction lamp post in the picture and at the same time producing an effect that just looks like the lamp exploding!

that's not unlikely

First of all, Wayne Pryde has been very low key about this. As far as we
know he has never claimed that it was a meteorite and he never claimed that
he saw anything. This is not really expected behaviour for a faker, because
their goal is to make the event more sensational.

But when someone would try to create a hoax image of a meteorite it seems very
unlikely that they would come up with an image like this one.

If he never claimed that it was a meteorite, why do you think he was thinking of one when he (if he) created it ?

Besides this world if full of strange people doing strange things that most people don't understand why.
And the internet is the perfect way for those people to pub those strange things.

[Doc Bluto]

Well no.. not at all. What I am saying is that this IS a factor that must be considered. And no, you cannot psychoanalyze me here... that would be as fruitless as your guesses about the "image in question". I will say that I am simply glad that the hoax possibility is being discussed. That's more important than what you might think. And, you are correct, this may be an image of something truly extraordinary. I have never denied that possibility. I am trying to demonstrate (as MrMoon has done with some degree of success) that intentional manipulation MUST be regarded as a prime candidate - regardless of your "feelings". And, that's a possiblity that has been intentionally ignored - despite the potential. That's it! That simple.

Doc Bluto. You seem to finally have come down to reason and engaged in a proper discussion, so I will attempt to talk to you and not treat you as a troll, but as far as I'm concerned this is a provisional state and your troll state may be re-flagged at the least show of misbehaviour.

Sorry Luis.. your attitude and demeanor is judgemental and that of an elitist. You're congratulating me because you think I've come down to reason? Your understanding of reason perhaps? Your are an arrogant one, aren't you? Don't level your petty little emotional judgments into my statements - you are the one I refer to when I mention those that don't tolerate it if other don't play by YOUR rules. You point your finger at me and state that I misbehave? You're an arrogant jerk.

[Doc Bluto]

You came late into the discussion. Within the first 10 pages of the thread it was understood that it was decided to trust the photographer. We decided to

YOU decided.. huh? This is EXACTLY what I mean by YOUR ego IN the WAY. You've proven my point to perfection sir. Well, I'm NOT playing by YOUR decreed rules bozo! You arrogant windbag!

[Doc Bluto]

Doc Bluto, now that we understand each other's point,

I wouldn't assume that.

Most of us posting theories started from the assumption that this was not a hoax. We full and well accepted the possibility but dismissed it.

So what's the point in anyone else playing your elaborate little guessing game? You've personally decided the rules then have you? Hmmmm.. I see. I see. So, apparently, you have discarded 1) Occams Razor, 2) scientific method 3) other view points. Again, for the 80th time, I am not fronting a theory, I am stating you cannot approach this with a predetermined result in mind. You can't 'decide up front that this is NOT a hoax'. Which is exactly what you have done. That's just plain wacky!

[Doc Bluto]

http://images.somethingawful.com/mjolni ... /Piney.jpg

Aside from EXIF data... which was intentionally removed, and the obvious sponsor - what happened here?

[Doc Bluto]

Here's a good challenge for you Luis...

http://www.godlikeproductions.com/galle ... category=7

take a peek... this will challenge you. Really, no kidding. Could it be real or not?

[victorengel]

I'm wondering if we can assume that the central part of the flashed insect image, the part that looks yellow, really is yellow. Even though someone hundreds of pages back showed that the flashed insect image is not saturated, can we assume that the response of the CCD or CMOS (whatever) chip is linear enough that it will faithfully capture the RGB components in the right ratio under all the lighting conditions in this image?

One potential issue is metamerism. Metamerism is an effect created when two items with different spectral properties appear to be the same color under one light source and different colors under another light source. In this case, the light sources are the ambient light from the sunset and the camera flash.

Another issue is white balance processed by the camera. The sunset was likely warmer (more red, less blue) than pictured. The camera flash, on the other hand, is always the same, within the precision of the recorded image, anyway. With the camera compensating for the warm sunset colors, anything illuminated by the flash would tend to look bluer than normal. The shift would tend to go in rainbow order, and standard color tools in graphic editing programs such as Photoshop show these colors in this order. An adjustment of the color using a hue adjustment tool can give some idea what the color would look like without the adjustment. An alternative would be to do an RGB adjustment, adding red and subtracting blue. I've done that here:


The object, then, if not yellow, might be a more orangy yellow. The camera's color adjustment, though, is not so extreme as to adjust the color beyond a range of yellows.

It's also possible that the blur from being out of focus effectively blends other colors together. Yellow is actually a composite of red and green. It's possible the insect was black with green and red spots and that the blur blended the two to look yellow. Since we have a before and after picture, though, I think we can safely assume that any brightness in the during picture came from the event. In the bug theory, that would mean any light in the during picture not present in the before and after originated from the camera's flash and was reflected off the insect. Here I show the insect in the hue and saturation we get without any adjustment.


I separated the picture into hue, saturation, and lightness channels and adjusted the lightness up so that the image wouldn't be too dark to see clearly. I made this adjustment only to the lightness channel, though, rather than using a simple histogram stretch operation, in order to preserve the color. So the color here should be reasonably accurate, but with the camera's color compensation applied. I've also shown the channels individually, first the lightness, then hue, then saturation. It should be noted that Jpeg compresses lightness the least and color information the most. This is quite evident when you separate the channels as I have done. This, of course, adds more uncertainty to the original color.

Here, I've adjusted the colors so that the putative specular reflections on the wings are close to neutral, in an attempt to undo the compensation made by the camera. I show the unadjusted and adjusted pictures next to each other.


You can see that the color of the body does not change that much. I think it's reasonably safe to assume the color shown is representative of the insect. As someone pointed out, though, it could be the reflection of the insect's eye, pollen, a yellow abdomen, or maybe something else we haven't thought of.

I'm thinking of flash photos of things like interiors of houses that you see, where there is something of a single colour (eg a wall), but parts of it are close to the camera and flash, whilst other parts are further away. As you go along the wall in the image, there is not only the expected change of brightness, but also a significant change of colour. From what I can remember, typically the further (and therefore fainter) part will look a lot greener. This is very obvious in emulsion type images, but it also happens in digital images as well.

In the example you mention, the nearer portion of the wall has a higher percentage of illumination by the flash, owing to its being closer to the flash. The more distant part of the subject is more influenced by the ambient light. If the ambient light is a different color than the flash (typical), then you will see a color shift. In the picture that we're studying, given the insect theory, the insect image is produced entirely by the flash (or at least insignificantly by anything else).

Finally, since most insects that I have seen are to some extent "shiny", I would expect that there would also be a specular reflection of the bluish-white flash that will "dilute" the true colour by some amount.

Yes. See my simulated image (described around page 86 or so) http://the-light.com/Photography/bugsimulation.jpg

In my simulation, the insect had a specular reflection on the head/thorax. Motion blur spread this specular reflection to make the horizontal white streak. The vertical white streak is imaged from the wing. My source picture, though, was illuminated not by flash but by available light, so the locations of the specular reflections would be different.

[victorengel]

http://images.somethingawful.com/mjolni ... /Piney.jpg

Aside from EXIF data... which was intentionally removed, and the obvious sponsor - what happened here?

Good grief! The motion blur is not even in the right direction. That's a very poorly done fake. What was the point of posting this, by the way?

[Anthonyg]

I think it is an insect that was flying between the lense and the sky.

[skyglow3]

Ok, in that terms it seem unlikely. but...

Taking a picture and being lucky cathcing a bug in such a way that it produces that trail wich happens to end just in the only malfunction lamp post in the picture and at the same time producing an effect that just looks like the lamp exploding!

that's not unlikely

First of all, Wayne Pryde has been very low key about this. As far as we
know he has never claimed that it was a meteorite and he never claimed that
he saw anything. This is not really expected behaviour for a faker, because
their goal is to make the event more sensational.

But when someone would try to create a hoax image of a meteorite it seems very
unlikely that they would come up with an image like this one.

If he never claimed that it was a meteorite, why do you think he was thinking of one when he (if he) created it ?

Besides this world if full of strange people doing strange things that most people don't understand why.
And the internet is the perfect way for those people to pub those strange things.

Well then, where should have the bug stopped in the picture? It's not even exactly on that lampost anyway. I think he could've made sure it lined up a bit more.

Well I haven't seen strange hoaxes or fakes like these, well maybe 1, but I can't remember. And as Ernst Lippe said the person would try to make the event more sensational. Ernst Lippe was considering the possibility of this being a strange hoax, and it seems to me to be low, and you actually agreed "Ok, in that terms it seem unlikely".

You say that it looks like a lamp exploding, wouldn't Pryde have even though of the fact that there would be damage? If he did think of it, but it was all part of his hoax plan, than that makes it more unlikely. The trail does not start from the edge of the picture, and I would think if he tried to convey the lamp exploding, he would have made the trail lighten smoothly, rather than abruptly stopping somewhere in the clouds. Also, why would he induce a slight curve or "wobble" into this trail? That wobble makes it that the chance of it being a shadow is nearly nil. So, lets say he didnt know that there was wobble, and that he simply tried to draw the streak with his mouse. But again why would he stop and not go all the way to the edge of the picture? Or lets say hes conscious that he put the wobble into the picture, why would he do that if he made it so the "flash" lines up roughly with thta not working light? THat seesm unlikely. If I were trying to make something that looked like a shadow I'd do it different.

trail wich happens to end just in the only malfunction lamp post in the picture and at the same time producing an effect that just looks like the lamp exploding!

Well, the effect would be caused by the bug being lit up by the flash, and the trail would stop at the bug because thats the end of the exposure.

skyglow1

[Tabu]

This is a very good example of how these things grow into legendary proportions.You take a very simple event like a bug passing in front of a camera in an interesting manner.Then you add layers of complexity to it till it is beyond the realm of reality.They did the same thing with crop circles.The reserches were finding all kinds of amazing propertys in them when all the time they were made with a board and sweat.the same goes for rods the speculation on them was intertaning to say the least.
fact is that this is a picture of a bug crossing in front of a camera nothing more.The odds of this happeing are very small I suppose but then the odds of winning the lottery are 13mill to one but people do it every week.