Strange streak discussion: 2004 Dec 7 APOD

Comments and questions about the APOD on the main view screen.
victorengel
Science Officer
Posts: 158
Joined: Tue Dec 07, 2004 11:29 pm

Post by victorengel » Tue Dec 14, 2004 7:58 am

DC wrote:I think it is good someone is trying to reproduce this effect with bugs, which I don't think will work.
Hmmm. I thought I did fairly well, considering I started out with an image of an insect that is most likely a different species, different size, different orientation, different speed, etc. It's a model, and models have limitations.
The image below is an invert of my previous double diff. I have labeled the image components the way I understand the bug theory. My concerns are:
Has it yet been established which direction the insect was traveling? I don't think so. The anatomy would be essentially reversed depending on when the strobe fired.
The wing/thorax refections are very narrow, which cannot be reproduced with a blur effect.
Let's separate the wing from the thorax. OK, we know insect wings are attached at the thorax, but we don't have distinct lines, so some of that is guesswork. The wings, though, extend beyond the reach of the body. Does anything else? Yes. Legs can, A long tail would do the same, as would extended antennae. I think the bluish wisps are probably reflections from the wings, but it's possible they are reflections from other body parts. See my post earlier where I show that one of the winglike shapes actually comes from the reflection of the head and thorax in my model.

Assuming what's pictured is wings, though, recall that we're talking about a period of time during which A) the insect is traveling through space ,B) the wings are moving relative to the body. and C) the wings are changing orientation. This last point is key. We're not simply seeing an illuminated wing. We're seeing specular reflections from the wing. Specular reflections only happen when the angle is just right. In the case of the head and thorax, the angle remains relatively constant, so the specular light trail follows the path of the insect. In the case of the wings, though, the location of the specular light depends upon the position of the insect AND the orientation of the wing. Thus it is possible for the apparent location of the specular reflection to be stationary or even move in the opposite direction as the insect. Obviously, I could not include this aspect into my model because I was essentially moving a bee without flapping its wings.
The wing reflections attach too neatly, with almost geometrical precision, to the thorax reflections.
I believe this is a consequence of the change in orientation of the wings. The left wing will sparkle at a different time than the right wing. Both will sparkle when they are facing the same direction, but this same direction happens at different phases of the flapping cycle. How far apart they are will depend upon the orientation of the bee to the flash and what the change in orientation is like of a wing during this kind of flight (I would expect a different pattern for hovering flight, which I think is very close to what my model was doing).
There is no room for the required connectional anatomy here.
There is, but you have to factor time into the equation. We are not looking at an instant in time. We are looking at a segment in time. I think the two wing images were made at different times and may even record one wing in its up position and the other in its down position while the insect is nevertheless horizontal. This is possible if the two wings are imaged out of phase from each other. Believe it or not, I don't have the time or equipment to conduct tests along these lines. If I did, I would.
Also, the thoraxic reflection should be a complete circle, which should not intersect the abdominal oval.
No. Remember, the insect is moving. At the instant the strobe starts, the rear end of the thorax is behind the position of the leading edge of the abdomen when the strobe finishes. How much overlap is related to the relative speed of the insect and the duration of the flash.

Guest

Re: Paintshop Pro's Solution

Post by Guest » Tue Dec 14, 2004 8:08 am

Anonymous wrote: As has been said before, the trail does appear to be straight but I think we're in agreement it's not really; rotating it to horizontal and squeezing left-to-right shows this fairly clearly.

Image
(click for larger version)

This also tends to argue against any high-energy explanation; contrails could wobble a bit as well, but then there are plenty of arguments against that hypothesis (it has to appear and disappear within 15 seconds, start and end within the frame yet be consistent in width and density despite touching the horizon - far too many coincidences for my taste, and that's ignoring the flash at the end).
I tried to duplicate your image. In my case, the trail seems to descend just at the end, which should be more consistent with a meteor.

Image

victorengel
Science Officer
Posts: 158
Joined: Tue Dec 07, 2004 11:29 pm

Post by victorengel » Tue Dec 14, 2004 8:13 am

victorengel wrote:I think the two wing images were made at different times and may even record one wing in its up position and the other in its down position while the insect is nevertheless horizontal.
This would require the strobe duration to be half a flapping cycle. If the insect were flapping at 160 Hz, then the flash would have to last 1/320 second. This would produce additional translation blur, so I acknowledge there is a problem here unless the insect has many more strokes than has bee so far proferred. Before you dismiss it, note that midges flap at about 60,000 Hz.

Guest

Re: SIMULATION

Post by Guest » Tue Dec 14, 2004 8:23 am

victorengel wrote: That resulted in this image.


http://the-light.com/Photography/bugsimulation.jpg


Also, please note that the results are dependent on my source picture for the bug. It's likely a better source picture would give better results. But this was the best picture I could find.
Nice one Victor.

I ran the simulation on your bug image. I wonder if you could overlay some of them as you did in the one above. I would like to see how they match.

Image
Image
Image
Image
Image

DC

Post by DC » Tue Dec 14, 2004 9:11 am

About whether the trail is straight, I think straightness might best be deterimined by a line down the middle of the trail. The roughness at the edges of the trail may just be the faint trail disappearing into the background noise.

About the bug images. I think the complexity required for the bug's behavior to produce the visible image is inconsistent with the stark simplicity of the image. The wing reflections possibly being produced by the same wing at different moments of a flap, producing reflections that exactly match the borders of the water in the image? Too improbable.

If I can come up with the kind of processed images I want by the end of the week I'll come back and post them. I'm hoping there is a way to pull just a little more information out of the pictures by using real numbers. But maybe nothing significant.

Pierre P.

Maybe the sun! or dust !!!

Post by Pierre P. » Tue Dec 14, 2004 9:47 am

If sun was in a good position, and a trail from a plane was in the way, it could have made a shadow... but for the flash, I have no idea...maybe coincidence...

and for the line, was this picture taken from a 35 mm ??? if so, maybe dust on the film or on obturator (sorry I`m French speaking) in french we say "obturateur"...so...it could have been some kind of dust on the film or in some part on the camera...

hopefully the answer will be given :-)

Pierre

Guest

Re: Paintshop Pro's Solution

Post by Guest » Tue Dec 14, 2004 11:17 am

FIRST REMARK:
if it was indeed a bug, you still don't know the orientation to the camera


Imagine: you look at a wall, and see a full wall, but if you look at the wall from the side of the you only see a line

same goes for a pole, you can see a line (the full pole), but if you look from above you only see a dot...

so what about an insect?
the "belly" of an insect reflects alot less than his "back"
if it's a bee you've got his paws and a black colour on the belly
but if you look at the back of the bee/bug you've got a more reflecting surface without paws


SECOND REMARK

it's very hard for me to explain because i don't speak english.
what if the bug was making a turn?

Image

you see?
to me it looks like the bug during flight is taking a sharp turn away from the camera, flying in the direction of the light pole/the road

you know what i'm saying

Depending of the spacial orientation of the bug you see something different

(sorry for making this even harder)

but there's a good chance that this was already said in the previous 90 pages ^^

Yutacan

Post by Yutacan » Tue Dec 14, 2004 12:01 pm

the post above was me... i forgot to fill in the name

so now you know who said it

Ruidh

Flash before or after?

Post by Ruidh » Tue Dec 14, 2004 2:44 pm

victorengel wrote:Has it yet been established which direction the insect was traveling? I don't think so. The anatomy would be essentially reversed depending on when the strobe fired.
One of the EXIM data interpreters said the flash went on in eye reduction mode. Eye reduction mode flashes the flash quickly several times before opening the shutter to get the irises of the subject to react and avoid the reflection off the back of the retina which gives the dreaded "red eye" effect. I think that strongly argues for a flash at the beginning of the exposure.

victorengel
Science Officer
Posts: 158
Joined: Tue Dec 07, 2004 11:29 pm

Post by victorengel » Tue Dec 14, 2004 3:38 pm

DC wrote: The wing reflections possibly being produced by the same wing at different moments of a flap, producing reflections that exactly match the borders of the water in the image? Too improbable.
What do you mean by "borders of the water"? I don't understand what you're alleging is matching. That one wing would reflect in the opposite position from the other wing is not improbable, actually. If you think about the position and orientation of the wings over an entire cycle, it makes perfect sense. In fact, it is much more likely the wings will reflect the same when opposite each other than when side by side. To reflect the same when side by side, they have to be coplanar. The main problem is the time it takes to get from one position to the other.

Bob Peterson
Ensign
Posts: 35
Joined: Sun Sep 19, 2004 10:47 am

Post by Bob Peterson » Tue Dec 14, 2004 4:04 pm

I'm back as promised. The pixel detective work is very interesting. I'm not convinced about the pixel interpetation on the horizon as it relates to the streak. My contention is that all these events i.e. the streak, the flash, and the smoke or smoke-like phenomema seemingly in the area of the post, are independent events, unrelated events. I thought for a little variety I'd put some DARK CONTRAILS up. Note in the text the word EVENING, as it relates to dark contrails.

http://www.geocities.com/milo702000/dark/dark.htm

Ruidh

Post by Ruidh » Tue Dec 14, 2004 4:26 pm

Bob Peterson wrote:I'm not convinced about the pixel interpetation on the horizon as it relates to the streak. My contention is that all these events i.e. the streak, the flash, and the smoke or smoke-like phenomema seemingly in the area of the post, are independent events, unrelated events. I thought for a little variety I'd put some DARK CONTRAILS up. Note in the text the word EVENING, as it relates to dark contrails.

http://www.geocities.com/milo702000/dark/dark.htm
Address the question: why does the "contrail" appear below the horizon? NONE of the pictures you linked to show a contrail below the horizon.

Yes, the contrails superficially look like the streak in the picture. But the geometry is all wrong. Ignoring the horizon problem which itself is enough to disprove a contril theory, the streak is not between the observer and the sun. The sun is way off to the right. How do you get a dark contrail out of that?

Guest

Post by Guest » Tue Dec 14, 2004 4:44 pm

flies generally don't fly about on their own. its interesting there wasn't 'trails' of others caught by the camera.

i still think it's just the lamp shorting and the digi camera artifacting at the large change in light levels.

jcomeau

strange streak

Post by jcomeau » Tue Dec 14, 2004 4:44 pm

Is it possible that what you see is an object falling out of the sky at a fast rate of speed to leave a trail but small enough not to make a splash?
The flash behind the lightpost is the point of impact.The smoke? around the lightpost could be an optical anomaly.Excuse me,I think I need to find my peril-sensitive sunglasses

Saabgod

Post by Saabgod » Tue Dec 14, 2004 5:32 pm

What we are witnessing is a light that is blown out and the long dark beam is the shadow from the flash the flash is not visible to our eye due to the sun off in the distance .... the metering system cant differentiate from the flash and the still brightness from the sun ...so we get a picture that shows more light than is what is really there hence the reason the shadow appears so bold on the left

Wolf Read

Australian "Meteor" Streak

Post by Wolf Read » Tue Dec 14, 2004 5:33 pm

If you look closely at the water, it looks like an irregular whitish line reaches from approximately in the vicinity of the pole, and out toward the horizon. How about a lightning bolt that travelled under the water from a far distance? Or, perhaps, this is the reflection of the lightning bolt on the surface of the water. With this explanation, the streak would be the light pole's shadow.

-best,

Wolf Read
Oregon Climate Service

Bob Peterson
Ensign
Posts: 35
Joined: Sun Sep 19, 2004 10:47 am

Post by Bob Peterson » Tue Dec 14, 2004 5:45 pm

Ruidh wrote:
Address the question: why does the "contrail" appear below the horizon? NONE of the pictures you linked to show a contrail below the horizon.

Yes, the contrails superficially look like the streak in the picture. But the geometry is all wrong. Ignoring the horizon problem which itself is enough to disprove a contril theory, the streak is not between the observer and the sun. The sun is way off to the right. How do you get a dark contrail out of that?

Answer to question 1=Horizonal pixal config subject to interpretation.
Answer to question 2=There's no direct sun along the streak path. Note the grey clouds.

zigiz

Post by zigiz » Tue Dec 14, 2004 6:13 pm

I've seen this photo few days ago in some of the Inet resources.
Guess what? It's the first photo that shows real METEOR collision with the pierse (or some lightning decoration on it)

Guest

Re: Strange streak discussion: 2004 Dec 7 APOD

Post by Guest » Tue Dec 14, 2004 6:31 pm

Anoter thougt...

1st. The strange shodow ist realy staight forward - now curves, etc.

2nd. From the beginning to the end it has the same "thickness"

3rd. It seems to have something to do with the bright light in the middle of the picture

Maybe this is an artefact of automatic imageprocessing used in "up to date" digital camaras which are trying to compensate the bright star effect by darkening the affected areas in the picture.

I guess the camara thougt it was fotografing right into the sun and tried to compensate...

What do you think?

Peter

Guest_itsabob

animiation of all three frames

Post by Guest_itsabob » Tue Dec 14, 2004 6:47 pm

Image

Here, you can see that the streak is in all three frames. You can also just make out its reflection in the water (this is very faint).

J Joy

Re: Strange streak discussion: 2004 Dec 7 APOD

Post by J Joy » Tue Dec 14, 2004 6:53 pm

Another tack on replicating the bud theory would be to take the same model camera, or similar, set it up at dusk on a tripod, set to same exposure, or close to, point it primarily at the sky and toss handfuls of insect size objects in front of the lens while exposing. If the bug theory is correct you should be able to record some streaks similar to the picture.

Also, still lookig for an answer to the following question:

I've scanned much of this discussion, but haven't seen whether or not there has been a follow up on the condition of the lamp. Is it known what the status is of the lamp? Was the bulb and or lens shattered? Or was it just inoperatable? Was there any damage to the lamp housing?

Thanks,

JJ

M. John Lalonde

Strange streak APOD picture Dec 7 2004

Post by M. John Lalonde » Tue Dec 14, 2004 6:55 pm

The photogropher mentioned the slow shutter speed 1/20 and medium aperture 5.6. We learned that he used the flash on the camera.

My suggestions is that it is simply a reflection from the camera's flash on a rain drop. The reflection has nothing to do with the lamp post.

The raindrop is merely somewhere in front of the camera. I suspect it is fairly close to the camera because the flash will lose its strength as it travels away from the camera, and because of the relative size of the reflection. The shape of the reflection suggests a sphere with a distorted halo, similar to our perception of rain in front a cars headlights.

The dark streak could be a shadow caused by the raindrop blocking light from the camera. Keep in mind that the dark streak can be as short as a few inches and does not have to be travelling across our field of view , i.e. from top left to bottom right or vice-versa. The shadow could be radiating from the reflection, either towards the viewer, or away from the viewer at any angle.

Furthermore, I believe the answer has to be rooted in 'common sense' science that can be deduced from the evidence that has been provided by APOD. I don't think APOD would want to be embroiled in controversial topics such as whether Mork was really from Ork, nor would they have the time or resources to source out additional evidence.

Nanoo Nanoo from Canada,

victorengel
Science Officer
Posts: 158
Joined: Tue Dec 07, 2004 11:29 pm

Re: animiation of all three frames

Post by victorengel » Tue Dec 14, 2004 7:23 pm

Guest_itsabob wrote:Image

Here, you can see that the streak is in all three frames. You can also just make out its reflection in the water (this is very faint).
No you can't. For those who can't tell, the event frame is in here twice, alternating with the before and after frames. Take out the event frames and you can't see a thing.
Image
Last edited by victorengel on Tue Dec 14, 2004 7:49 pm, edited 1 time in total.

Bob Peterson
Ensign
Posts: 35
Joined: Sun Sep 19, 2004 10:47 am

Post by Bob Peterson » Tue Dec 14, 2004 7:47 pm

RE the animation. Uhmmmm. That motion I know I'm familiar with, but I can't quite nail it down. Oh well, s@#$% it.

JerWah

Re: Paintshop Pro's Solution

Post by JerWah » Tue Dec 14, 2004 8:10 pm

Anonymous wrote:
I tried to duplicate your image. In my case, the trail seems to descend just at the end, which should be more consistent with a meteor.

Image
I don't follow your logic that the dip at the end is "more" consistent with a meteor. To me it only eliminates a shadow or other illumination based phenomenom(sp?) (Doubly so given the fact that the shadow doesn't extend off the frame).

One of the strongest arguments against the bug proposed initially is that the streak was "too straight". At least to me the fact that there is any deviation in the line only adds additional credence to the bug theory rather than diminishing it.