Strange streak discussion: 2004 Dec 7 APOD

[Guest]

What if it is an artifact of the camera focal plane shutter. Does this camera have a vertical travel curtin or a horizontal travel curtin?

It uses a leaf shutter like 99% of all consumer point and shoot digital cameras.

[Guest]

The arc of the dark line intersects precisely with the light at the top of the lamp post, but the whitish line, which is also centered on the lamp and extends across the water, does not have the dark line as its origin.

The dark streak ends before the edge of the frame.

I think that the lamp flashed, possibly by the filament or tube arcing momentarily, causing a flash, which could have high intensity.

The dark streak is a camera artifact.

I do not think that this is an inbound event, rather the result of a momentary event at the pole. If the observer reports that the light was off, how is it that the lamp shows an intense incandescence? A momentary flash would be consistent with the observed effect. The rest is down to the vagaries of ccd camera recording?

Just a thought from a complete novice.

CCD artifacts with standard matrix sensors tend to only happen in horizontal or vertical directions. The camera in question (a Canon Powershot G2)has a standard matrix sensor, so nothing about the sensor would yield an artifact like this, unless someone had mounted the camera on an angle. But then the dimensions of the image would not match the actual image, because the rotated image would have to be cropped to yield something that looked level. To get it back to the proper size, they would have needed to resample the image, which would have been obvious to detect.

So, in other words, no.

[Guest]

The Dark streak cannot be a contrail, projectile, or a camera artifact. Contrails are wispy, this is a straight blur. Projectiles must arc, but this is a PERFECTLY straight line. If it were an overexposure due to a camera artifact, it would be horizontal or vertical since this is a digital camera NOT DIAGONAL. It must be a bug because it is a straight line and whether the insect is flying in a curve or not, in 1/20 of a second, it will appear to fly in a straight line. It is not totally a blur and only really appears at the end. Some use this to discount the bug theory. But from experience, I can tell you that when something is moving faster than the shutter speed/collection time of the camera, when the flash occurs it will leave an image and when the flash is not hitting it, it will be more of a blur. These pictures with a blur and then the "ghost image" are very very common.

[freeman]

Why is there so little mention about the top of the light pole--the glare and the smoky looking perimeter?

Am I wrong in thinking they are part of the event?

I see an event that includes a discharge of light and smokiness around the top of the light pole.

The question then would seem, what such event do little more damage than possibly the busted filament, would result in the relatively straight dark streak arching to the sky, and would not contradict experts who nix meteors and lightening.

It looks a little stormy in the background. Were there cumulonimbus clouds or conditions typical to a weather inversion?

Maybe the edge of a small short downburst could have contained particles and condensation. The edge might look straight and darker when veiwed latitudinally, lighter and more erratically shaped where the perspective changes, or where it reaches the equally cooled air near the lake surface.

Sudden pressure might have been enough to rattle the filament loose, causing a flash, without bursting the lamp.

--
freeman

[Jim]

Using Photo Shop to compare the color of the light pole "glint (flash)" which appears to come from a light pole with the color of the clouds directly above. It appears the colors are just about the same. It is possible the glint could come from a shinny object (polished top of the lamp, but in a salt water area I doubt there is much of any thing shinny). My best guess: Flash is a reflection off the water because the cloud is quite bright and the color values are close. The flash off the water just happens to match the light pole.

I don't think a burned out bulb flash would have such a close match to the cloud's color. Also, if the bulb was in a light pole fixture, it seems to me the flash would be downward. But, not bright enough to over load the cam.

The streak, could be a guy wire or hair. I am still stuck as to it's cause. Tinkering???? hummm

[Guest]

To add to my earler comments doubting the fly theory. (straight line, dark trail, ...) does anyone think a bug flying in such a straight line is going to be flying sideways so that you'll see a top (or bottom) view from the cameras viewpoint? Doubt it.

I do. Sideways is a relative term -- there are a number of directions and orientations that an insect could have been flying to look like that. And that shape could just as easily be a longer insect that is foreshortened by perspective. In other words, we may not even be looking at its "top" or "bottom" view. It could be the "behind" view.

[Guest]

This really looks like his CCD is malfunctioning on his imaging. It happens when you are around ocean water. The salt is in the air and it starts to corrode very fast within hours. The other is the constant pink dot in the water in the same location in all three photos its in the lower left of the picture in the water. There are several others in the photo. This is another tell tell sign of corrosion.That shadow up top is a flare caused by the CCD and its flaw I have had it happen to me many times with several cheap digital cameras such as Canon and all so Epson. Another culprit is batteries going dead. With digital its either on or off when a battery starts to die it can cause wierd things to happen to digital photos as most of us that use them know.

You were doing real good right up until that paragraph.

Salt air does not start a CCD "corroding" within a few hours. That is wrong

Hot pixels are not a sign that the CCD is corroding or about to fail. Pixels fail for a large variety of reasons. Yes, I agree that the CCD does have hox pixels, but it is not in any way related to the sea air. I have seen CCDs on cameras right out of the box, the very first shot, exhibit hot pixels. The CCD is simply defective in that one area. It's silicon and therefore can never be 100% perfect. The hox pixel is mapped out in software and you never see it again.

The shadow is not flare. If you look up in any photography text what flare is you'll see how wrong you are. What you may be thinking of is "blooming" something related to the overloading of pixels by too much light. Blooming can only occur in the verticle and in the horizontal, but not in a diagnol. IT's the inherent design of the CCD that prevents this. Also, the Canon CCD's do not bloom due to the anti-blooming gates built into the CCD chip.

All in all, IF and this is a big if, IF the image is not faked, the bug/fly explanation is the perfect fit. All other explanations fail to account for all the factors. The fly solution is the most simple and the most obvious one.

[davec@alum.mit.edu]

It would be interesting to have someone who has the same model of camera comment about the consistancy of the EXIF data his camera produces. I've got a different brand of digital camera, but it often records the flash going off when it didn't.

It seems odd to me that a person shooting a series of photo's at 1 every 15 seconds would have the flash on without a good reason.

-davec

[DC]

I see no anomaly in the timing of the images, other than reversal of before and after. The first image was taken at 6:52:37, according to earlier posts. The next was at 6:52:52. According to my calculator, this is exactly 15 seconds after the first image. The last image was taken at 6:53:07, which again according to my calculator is exactly 15 seconds after the middle image. This amazingly coincides with the photographer stating the images were taken 15 seconds apart.

[gg]

[Guest]

Canons actually fire their flashes at the end of the exposure, so the bug was headed toward the center of the image rather than away, but you have the right idea.

No they do not.

Actually they do. Unless I was on some sort of hallucinatory drug earlier today when I tested mine and reproduced that very effect.

[phule]

Canons actually fire their flashes at the end of the exposure, so the bug was headed toward the center of the image rather than away, but you have the right idea.

No they do not.

Actually they do. Unless I was on some sort of hallucinatory drug earlier today when I tested mine and reproduced that very effect.

The flash fires at the beginning of the exposure. I have used 10 different digital cameras (point and shoots) in the past 2 years. (Various friends cameras that I've borrowed). Every single one of them shot the flash at the beginning of the exposure.


Test it again.

[phule]

Canons actually fire their flashes at the end of the exposure, so the bug was headed toward the center of the image rather than away, but you have the right idea.

No they do not.

Actually they do. Unless I was on some sort of hallucinatory drug earlier today when I tested mine and reproduced that very effect.

And b.t.w I'm sitting here with a Canon A80 right now. The shutter is set to 1 second. The flash fires and /then/ the shutter closes. I can hear it. If I turn off mute, the camera plays the shutter sound at the /end/ of the exposure 1 second after the flash went off.

[James]

My Cannon 300D fires its flash at the beginning of the exposure.

[rxd]

I partly agree with schulk88@hotmail.com, the shadow path which was suppose to show a beam of light was due to the present weather in the area. If there are data showing if there are any aircraft or object in that area on that specific time maybe the would justify something. I have doubts the group who has taken this picture might have had a better idea of that area and why
the shots were clearly focus right on the center of what's going to happen.

[Ken]

Its a firefly zipping past very close to the lens. Perhaps caught in a gust of wind.

The firefly is too close to be in focus. It coincedently flashed its light while it was nearly inline with the little light pole. For the rest of its path ... from the upper left to the lower right, it was blocking some light, leaving what you think is a dark trail through the sky.

If it passed very close to the lens, say within a centimeter or 2, then in 1/20th of a second it could easily travel 2cm or so across the view at a speed of only 0.4m/s, appearing to streak across the sky.

Ionization traces preceding lightning travel upwards in an irregular path, directly away from the charged ground beneath.

I'm also in Vancouver, and watch the planes coming in often and have never seen contrail shadows looking like this ... and when was the last time you saw a jetliner fly under a thunder head, instead of way above, or moving around it? ... or a contrail that could be generated and disappear over a 30 second period while the clouds around hardly move?

meteor? why not? the white cloud to the right of the light could be a shockwave ahead of it ... but why would the meteor only be bright right at that moment instead of over the entire path, given that the light pole was not struck by it?

Who has some software that can take the suspect photo and the before or after photos and do a change detection? ... would provide a clearer view of the feature without the distracting background.



Ken Lord
Vancouver BC
kenlord [at] gmail.com

[Röd Vardogr]

I saw a similar "light" effect, it was scary. I was at the roof watching the stars and then... kind of a laser beam effect, I don't know came down from somewhere in the sky, it was a pale violet light but you could see it because of the blackened sky. That beam, or light pattern moved left-right-up-down and semi circles then it disappeared. It was so shocking that sometimes I'm afraid of goin out and stare at the sky.

[sparks]

Would you please post the images acquired immediately before and after the frame with the streak in it? A comparison might help to make it more obvious what's happening.

What is the distance to the objects in question? It looks as if the streak may be several feet wide in the center of the photo and tapers towards both ends. It should be possible to estimate the speed of whatever caused the streak if the length of the streak could be estimated.

The streak seems to have dissapated within a fifteen second time interval? Was the day windy or still?

[Guest]

Would you please post the images acquired immediately before and after the frame with the streak in it? A comparison might help to make it more obvious what's happening.

What is the distance to the objects in question? It looks as if the streak may be several feet wide in the center of the photo and tapers towards both ends. It should be possible to estimate the speed of whatever caused the streak if the length of the streak could be estimated.

The streak seems to have dissapated within a fifteen second time interval? Was the day windy or still?

Any reason you missed the links in the APOD?

http://antwrp.gsfc.nasa.gov/apod/ap041207.html

[Guest]

No they do not.

Actually they do. Unless I was on some sort of hallucinatory drug earlier today when I tested mine and reproduced that very effect.

And b.t.w I'm sitting here with a Canon A80 right now. The shutter is set to 1 second. The flash fires and /then/ the shutter closes. I can hear it. If I turn off mute, the camera plays the shutter sound at the /end/ of the exposure 1 second after the flash went off.

It depends on what mode you are using. If you are using the normal flash mode (or flash sync) it will fire at the beginning. If you use the special nighttime assist mode (usually indicated by an icon with a person with a star behind them) it will open the shutter first and then flash at the end. This is probably because if you have motion blur, you want the blur to trail the object rather than making it look like everything moved backwards.

I tested this on a Canon Powershot S500.

[Yutacan]

in the three images you can clearly see a pink dot in de water before the road


it's on all three images

i'm not a photograph expert, but what is that? a digital imaging artefact??



and, I red this whole thing through, and to me, it looks very possible that it's a firefly that flew by the lens
the flash has "wings" of some sort, you know
a fly has semitransparent wings doesn't it?

[Mark Tranchant]

I expect someone has already suggested this, but here's my solution (also posted on Slashdot):

The light levels are fairly low: the EXIF data from the big image reveals that the Powershot G3 used 1/20s exposure at f/5.6. I reckon the streak and the blur are very, very close to the camera, and that the intersection with the streetlamp is conincidence.

I believe that the mystery object is an insect flying "north-west" (i.e. towards the top left of the camera). The EXIF data tells us that the flash was fired, although goodness knows why any decent photographer would use a flash for that shot.

The flash on most cameras fires at the beginning of the exposure time, and the insect is captured in flight and out of focus near the middle of the frame. It then continues flying for the rest of the 1/20s exposure causing the black streak.

Where do I go to collect my prize?

mark@tranchant.plus.com

[Guest]

This takes one of the before/after pictures and attempts to amplify the color change between that frame and the central frame. Therefore the ocean is getting more red as the color of the reflected sunlight changes. (I think I used the before picture, but I'm not sure)

I think this effectively shows that what I call the reflection in the water is chromatically unrelated to anything else in the frame, such as in the clouds.

[Guest]

it sure looks like a hair with the follicle end refracting light.

imho...

[DC]

I also think the picture I just posted shows there is enough of a color difference between the flare and the reflection that they are unlikely the result of a flash reflecting off an insect.

I don't know if this is exactly what Ken Lord wanted.