What is this streak of light caused by?

Peter Stewart

New Member
IMG_7220.JPG IMG_7221.JPG IMG_7222.JPG

This is yet another photographic anomaly that I can't explain...this is night shot taken through my bedroom window using the window frame as steadying support and shot with no lights on around midnight.
So please can someone explain what the streak of light is caused by at the top of the shot?
It had just snowed. The sky was steel grey and still full snow. There shouldn't be any way that this flash of light appears.
 
Last edited by a moderator:
Probably a reflection, as it appears in the same place in multiple both photos.
 
Last edited:
Probably a reflection, as it appears in the same place in multiple photos.
A reflection on the window?
It does look sort of like a reflection from a ceiling track light, but the light appears to stay in the same spot relative to the building in each photo. If I put the edge of a ruler along the line of the light it hits the edge of the tower with the 3 protrusions at the same spot in each photo. But the last two photos show more of the left side of the building. If it were a reflection on the window, wouldn't the light appear farther right than it does in the last 2 photos?
 
Last edited:
If it were a reflection on the window, wouldn't the light appear farther right than it does in the last 2 photos?
Not, if the camera was just turned a little, but not moved horizontally, e.g., on a tripod. I've made a "stereo pair" of the two images, one with the least of the left side of the building, the other with the most. There is no stereo effect in any of the two combinations (first image - left, second - right, and vice versa).
 
A reflection on the window?
It does look sort of like a reflection from a ceiling track light, but the light appears to stay in the same spot relative to the building in each photo. If I put the edge of a ruler along the line of the light it hits the edge of the tower with the 3 protrusions at the same spot in each photo. But the last two photos show more of the left side of the building. If it were a reflection on the window, wouldn't the light appear farther right than it does in the last 2 photos?

I took these shots - David Beckham used to live in an apartment, the one with the tower at the far end of the building in the photos..until Posh Spice got hold of him and ruined his life. But that's another story - anyway for the record, there were NO lights on. This is not glare or lens flare, both of which I'm familiar with. It's something external.
Also if you zoom in you will see several tiny white dots, which you guys will immediately say: it's a blown pixel.
Well it isn't. I took other shots that night and none show the same tiny aberrations as this set.
It was still snowing as you can see in some shots and the sky was laden with more snow to come.
No lights on anywhere other than on the building itself. Hope this helps
 
Also if you zoom in you will see several tiny white dots, which you guys will immediately say: it's a blown pixel.
Well it isn't. I took other shots that night and none show the same tiny aberrations as this set.
I certainly see some white dots which do look like stuck pixels, as they are in the same place in each shot. (By which I mean the exact same co-ordinates in the frame, regardless of the movement of the subject matter.)

The white streak I am not sure about. It appears to have a double image to it, which I thought might be a sign of a reflection in double glazing, but the rest of the image also has that doubled appearance which suggests it is just caused camera movement.
 
By the way, looking at the five photos you posted, the first three are identical (all have the same file name too). The fourth and fifth also appear to be identical to each other with the exception of one being lightened. Even the streaks of falling snow are identical from frame to frame.

[compare]
4.JPG 5.JPG
[/compare]
 
By the way, looking at the five photos you posted, the first three are identical (all have the same file name too). The fourth and fifth also appear to be identical to each other with the exception of one being lightened. Even the streaks of falling snow are identical from frame to frame.

[compare]
4.JPG 5.JPG
[/compare]

I've removed the duplicates in the OP. There's three images, 7220,7221 and 7222

22 is a brighter version of 21 - possibly the camera was set to HDR.

There are stuck pixels. They won't always show up in shorter exposures - not exactly "stuck", just more prone to going white.

@Peter Stewart, can you email me the originals, metabunk@gmail.com
 
If you line up the buildings between the two photos, the streak is perfectly aligned, but longer in the second image. It's also more blurred in the second image, just like the buildings are.

[compare] pic1.jpg pic2.jpg [/compare]
 
Last edited:
Slight possibility that it's a wire, like a phone line. We from the rain/snow and reflecting the lower light
 
Slight possibility that it's a wire, like a phone line. We from the rain/snow and reflecting the lower light

Were there any astronomical events the night the images were taken? Im not an expert but to me it looks like a meteor streak... the only reason Im even suggesting that is because it looks like there are a few stars visible.. unless those are just snow flakes caught by the exposure.

@Peter Stewart What was your exposure set to and the exposure time?
 
Were there any astronomical events the night the images were taken? Im not an expert but to me it looks like a meteor streak... the only reason Im even suggesting that is because it looks like there are a few stars visible.. unless those are just snow flakes caught by the exposure.
The bright spots that look like stars are just stuck pixels. They are also present on top of the buildings and trees, and are in the same position on the frame in both pictures.

upload_2017-3-30_20-2-38.png


The sky looks totally overcast, and as Peter said snow was falling I don't see how a meteor could be visible. Especially as the upper end of the streak is visible in the same place in consecutive photos. If it was a moving object then the streak could not occupy the same place in consecutive shots: it would be different sections of the same line.
 
IMG_7220.JPG IMG_7221.JPG IMG_7222.JPG

This is yet another photographic anomaly that I can't explain...this is night shot taken through my bedroom window using the window frame as steadying support and shot with no lights on around midnight.
So please can someone explain what the streak of light is caused by at the top of the shot?
It had just snowed. The sky was steel grey and still full snow. There shouldn't be any way that this flash of light appears.
Some answers here reckon it's some reflection of a light, strip light suggested too. Well, for absolute sure, there were no lights on in my bedroom when these shots were taken,
And for the record, I don't have one single strip light
 
Some answers here reckon it's some reflection of a light, strip light suggested too. Well, for absolute sure, there were no lights on in my bedroom when these shots were taken,
And for the record, I don't have one single strip light
And is there any kind of wire attached to the house (phone, electric) that light could be shining off?

Do you know the time interval between the two shots, and the length of exposure?
 
Were there any astronomical events the night the images were taken? Im not an expert but to me it looks like a meteor streak... the only reason Im even suggesting that is because it looks like there are a few stars visible.. unless those are just snow flakes caught by the exposure.

@Peter Stewart What was your exposure set to and the exposure time?
I'll have to get back to with exposure details, but those stars/snowflakes you refer to are something else.
They're not blown pixels as you'll see the 'stars' appear in different spots and all other shots taken that night in different directions there are not one single similar 'star' . And they aren't snow flakes...zoom right in...they're perfect spheres so not snowflakes. I know what they aren't. Just having a problem with what they are!
 
I'll have to get back to with exposure details, but those stars/snowflakes you refer to are something else.
They're not blown pixels as you'll see the 'stars' appear in different spots and all other shots taken that night in different directions there are not one single similar 'star' . And they aren't snow flakes...zoom right in...they're perfect spheres so not snowflakes. I know what they aren't. Just having a problem with what they are!
Ps. Appreciate your input.
 
I'll have to get back to with exposure details, but those stars/snowflakes you refer to are something else.
They're not blown pixels as you'll see the 'stars' appear in different spots and all other shots taken that night in different directions there are not one single similar 'star' . And they aren't snow flakes...zoom right in...they're perfect spheres so not snowflakes. I know what they aren't. Just having a problem with what they are!

If you are talking about these white spots then they are in the same place in the different shots you have posted. Actually they all seem to be shifted a couple of pixels up and right in the second image, but their relative positions are identical:



Those white dots are a camera artifact - they don't change position even though the picture is framed differently. But the white streak is different - it moves precisely with the subject matter. To me that suggests that it is a physical object and also at a similar distance to the house. If it was a reflection, and/or a long way closer or further to the camera than the rooftop then you wouldn't expect it to maintain its exact relationship to the house.
 
Last edited:
I don't think it's a moonbeam (the sky is overcast) but I do wonder if it is related to this light over here. There is a faint pattern of bands in the sky at the top right that seems to radiate out from this light source too.


upload_2017-3-30_21-6-47.png
 
There is a faint pattern of bands
yea it looks like an old telephone cord :) I'm having a hard time placing the lights around the house (which may explain the dark left corner) because of the shadows of the trees, lighting on the front of the house and whatnot.

Perhaps since peter still lives there (?) he can share some new photos of the house at night that better show the manmade lighting.
 
The bright spots that look like stars are just stuck pixels. They are also present on top of the buildings and trees, and are in the same position on the frame in both pictures.

upload_2017-3-30_20-2-38.png


The sky looks totally overcast, and as Peter said snow was falling I don't see how a meteor could be visible. Especially as the upper end of the streak is visible in the same place in consecutive photos. If it was a moving object then the streak could not occupy the same place in consecutive shots: it would be different sections of the same line.

Ahh ok. The pixels looked like stars to me, and the way the image was shot it reminded me of how dark winter nights appear sometimes while I was living in Canada... even on the darkest, overcast and coldest of nights, with snow falling, I could sometimes spot a star or two. Makes sense now though (also referring to your diagram with the lines).
 
The bright spots that look like stars are just stuck pixels. They are also present on top of the buildings and trees, and are in the same position on the frame in both pictures.

upload_2017-3-30_20-2-38.png


The sky looks totally overcast, and as Peter said snow was falling I don't see how a meteor could be visible. Especially as the upper end of the streak is visible in the same place in consecutive photos. If it was a moving object then the streak could not occupy the same place in consecutive shots: it would be different sections of the same line.
Two of the photos are the same shot, my mistake uploading them. They're honestly not stuck pixels which on zooming in would be square shaped. These are perfect spheres and are in different places on the photos that are slightly different.
I've taken about 10,000 photos with this canon eos5 mark2 and NEVEN seen one single stuck pixel in any of those 10,000 shots. So probability says...
 
IMG_7235.JPG
yea it looks like an old telephone cord :) I'm having a hard time placing the lights around the house (which may explain the dark left corner) because of the shadows of the trees, lighting on the front of the house and whatnot.

Perhaps since peter still lives there (?) he can share some new photos of the house at night that better show the manmade lighting.
Ok. I can do that. The only light is that thing stuck to the wall on left of this photo. In this shot the porch light is on at the apartment fronting the building. In the other shots it isn't.
 
THis looks like it was taken from a slightly different position to the other images.
Looks to line up pretty well to me.

And by overlaying that image you can see where the centre of the light is. The streak seems to point almost directly to it. As that's the main light source in the image, I am leaning towards it having something to do with that light on the side of the house.

upload_2017-3-30_22-8-40.png
 
If you are talking about these white spots then they are in the same place in the different shots you have posted. Actually they all seem to be shifted a couple of pixels up and right in the second image, but their relative positions are identical:

[compare]circle2.jpg
circle1.jpg
[/compare]

Those white dots are a camera artifact - they don't change position even though the picture is framed differently.

Yes, the white dots are definitely a camera artifact. "hot pixels" which show up white or brightly colored in long exposures. They do not line up exactly because the images in the OP are cropped and rotated slightly - probably just to straighten it. But comparing it to a raw mage Peter sent there's that specific pattern of dots on the right of the sensor, highlighted here:
20170330-144930-gfrh6.jpg

With the same pattern of dots in the same location in another photo that is looking down:

20170330-145153-8uv7q.jpg
I think the best approach to solving what the streak of light is would be to try to duplicate the shots as closely as possible.
 
I captured a somewhat similar streak, while filming a video through the glass pane:
Screen Shot 2017-03-31 at 00.00.11.png
The streak was due to a linear scratch or a strip of detergent on the pane, illuminated by a bright outside source, in this case, the reflection of Sun from the aircraft.
 
It is lens flare. Flare radiates from a bright point of light. Flare is often more prominent the further toward the edge of the frame the point of light is. The difference in sharpness between the streaks and the building point to in-camera creation. The streak is brighter when more of the light is in the frame, pointing to lens flare. The darkness in the upper corners (framing the large circular shaped lens flare) moved with the camera, precluding vignetting by lens hood or wide open lens aperture. Zoom lenses tend to exhibit greater flare than prime (fixed length) lenses due to having more lens elements.
 
It is lens flare. Flare radiates from a bright point of light. Flare is often more prominent the further toward the edge of the frame the point of light is. The difference in sharpness between the streaks and the building point to in-camera creation. The streak is brighter when more of the light is in the frame, pointing to lens flare. The darkness in the upper corners (framing the large circular shaped lens flare) moved with the camera, precluding vignetting by lens hood or wide open lens aperture. Zoom lenses tend to exhibit greater flare than prime (fixed length) lenses due to having more lens elements.

Seems a bit odd to have just this one solitary streak of light though. To me it looks more like the light smear shown by trailspotter.

Another clue from a photo that Peter sent me. The pattern in the upper right corner in the "streak" photos was thought to be related to the light. But in the new photo, in landscape orientation (the upper photo below), the pattern is more defined and appears to be something in front of the camera.

20170331-061515-n8rh3.jpg

I think the only way to solve this for good would be to replicate the photo.
 
Lens flare is also more typically oriented roughly through the center of the image. This is very off center.
 
Another clue from a photo that Peter sent me. The pattern in the upper right corner in the "streak" photos was thought to be related to the light. But in the new photo, in landscape orientation (the upper photo below), the pattern is more defined and appears to be something in front of the camera
Spider's web close to the window, casting a shadow upwards away from the light and with one filament catching the light?

But I'm still puzzled by the way that the streak is in different places in the frame between the two photos, but is in exactly the same position relative to the distant building. If it was something close to the camera then a small shift in camera position would make it appear in a very different place relative to the background.
 
In post 30 tree detail appears physically obstructed in the upper right corner. AND we have not seen what the length of exposures were, and what ISO was used. The Canon EOS 5D MKII has a long exposure issue with heat affecting the sensor, sometimes causing 'purple corners'. The OP could take a series of exposures with the cap on in a dark room, starting at the same exposure and ISO, and doubling the time for each successive shot for 6 shots. Then take the resulting files and equalize or auto-level them. This might show or rule out a problem with the camera.
 
Back
Top