A Vision Question.

July 19, 2005 4:57 AM

When I look at something (without my glasses) that's about 100 feet away, it looks a little blurry. If I look at a photo that includes stuff that's 100 feet away from the perspective, it's not blurry, of course, it's just smaller. BUT, if I look in a mirror, let's say a rear view mirror of a car, at something that's 100 feet behind me, it looks blurry, rather than just small.

So why doesn't a mirror act like a photo and just make me see the thing smaller? The plane on which I'm seeing the object, after all, is just a few feet away, same as a photo. How does a mirror actually maintain the depth of field, but a photo doesn't?

(Any auto accidents resulting from Me-Fi'tes playing with this optical trick themselves on their way to work this morning are not my responsibility.)

posted by Framer to health & fitness (20 comments total)

Put simply - because you aren't looking at the mirror. You are looking through the mirror at a point that is pretending to be behind it. As far as your eyes are concerned, the mirror is as substancial as a pane of glass looking into another world.
posted by twine42 at 5:00 AM on July 19, 2005

The photons reflecting off the subject are still travelling a long distance, whether you look at the subject directly or through the reflection from a rear-view mirror. Your eyes and brain resolve the image data accordingly.
posted by Rothko at 5:08 AM on July 19, 2005

The above two answers are correct. If you had a video camera installed so that your rear-view mirror was in fact a TV screen, you'd be able to see 100 feet away without it being blurry.
posted by Mwongozi at 5:13 AM on July 19, 2005

How far the photons travel doesn't matter, they don't have odometers. Simply put, the point is that looking at a point p on a photograph from two slightly different angles will result in seeing the point the same colour, whereas looking at a point p on a mirror from two slightly different angles will result in seeing different colours (ignoring coincidences therein).
posted by fvw at 5:15 AM on July 19, 2005

IANAR (Anything relevant): However, it seems to me that when you're looking at the picture you're looking through 2 lenses. First you're looking through the lens in your own eye, and second you're looking through the camera lense that was used to take the picture.

The camera lens used to take the picture is designed to make objects 100 feet away non-blurry when seen through a typical human-eye lens. If it weren't, the picture would also look blurry.

When you look at objects 100 feet away in the mirror, you are missing that second lens. You could add the second lens by taking a telescope or binoculars and looking in the mirror with that. (I don't actually know if that would work, but it would be interesting to try!).
posted by duck at 5:16 AM on July 19, 2005

How far the photons travel doesn't matter, they don't have odometers.

How far away the subject is changes where the light falls on the lens in your eye, and accordingly will change the focal point on the back of your retina and the sharpness at which your brain will register an image of that subject.
posted by Rothko at 5:21 AM on July 19, 2005

there are things called camera obscuras, which are rooms with a tiny window. light comes in through the window and makes an image on the wall. that's an image, and would work as you expect - you would see things in focus, even though they were far away. same with a film projector. in each case, you are looking at an image on a surface.

a mirror (or rather, what you see in an image) is not an image (not in the same way). you can tell the difference because when you look at a photo, or at the projection in a camera obscura, and you move, the image stays "the same". but what you see in a mirror moves as you move.

i tried writing more to explain the difference (it depends on whether light is scattered or reflected from the surface you are looking at), but it got too complicated. i can't think of a good clear explanation that's simple enough to write in a few paragraphs. sorry!
posted by andrew cooke at 6:31 AM on July 19, 2005

how about this - building on what i said above about the image not moving if it's a "real" image, but what you see in a mirror moving.

an image - a photograph, or a projection on a wall - is there. it exists on the surface in an unchanging way. how it appears doesn't change when you move your head - objects don't move relative to each other - because it's "fixed". we can describe what happens using simply the light from the surface and your eyes.

a mirror is different, because it's part of a larger system. we can't describe what you see without including your eyes, the mirror, and what you are actually looking at. the mirror is just part of a larger process.

what you ask is a good question. i think you could answer it in terms of coherence of light. that's quite a "deep" physical concept.

a mirror is "mirror-like" (shiny). it preserves the coherence of light waves as they are reflected. a piece of paper with something printed on, or a wall with something projected on it, is rough. it doesn't preserve that coherence. that lets you draw a line in your explanation - you can stop at that point without worrying about previous things.

ha. i'm sure that makes no sense at all, but it is an interesting question.

anyway, if you buy that explanation on some level then you can see why there's a difference. on one case we can explain things using just nearby objects (the paper, the wall, your eyes). in the other, we have to use the larger system, which includes things further away. they're completely different, optically, and so behave differently, even though (in a "physics naive" sense) they "look the same".
posted by andrew cooke at 6:41 AM on July 19, 2005

completely different approach:

when a physicist thinks of an image, they always think of it existing in a certain plane. for a painted picture, that plane is obvious - it's the sheet of the canvas.

the image you see in a mirror, just like the image you see in other optical systems (binoculars, camera, telescopes), has a plane too, but that plane is not at the same point as the mirror.

a mirror is much more like a lens than it is like a piece of paper.

take binoculars. when you look through binoculars, you see an image that is in a certain plane. whatever that plane is, it's certainly not the lens just next to your eyeballs! you are not looking "at" the (surface of) the lens, but looking "into" it. the same with a mirror. you are looking "into" the miror, just like you look "into" a lens. or looking through.

mirrors are like lenses. you look through them. the (plane of the) image you see is further away, and so is blurred. in contrast, a photograph you look at.

that sounds like a simple difference - just a small change in words - but the physics is completely different in the two cases.
posted by andrew cooke at 6:48 AM on July 19, 2005

a mirror (or rather, what you see in an image)

two posts up. should be: "a mirror (or rather, what you see in a mirror)".

sorry.
posted by andrew cooke at 6:49 AM on July 19, 2005

Also note that when you look through a mirror, you maintain depth perception due to the fact that stereoscopic vision is preserved. This is because your eyes have slightly different views of what you're observing, even through a mirror. Like twine42 said, viewing an object in a mirror is the same as viewing an object that's just as far behind it. Light is simply bounced off of the mirror, not projected onto it, so you still have to focus on faraway objects just as if you viewed them directly. Of course, convex and concave mirrors complicate the issue by causing the light to converge or diverge after bouncing off of it (either creating a wide-angle view, like the "objects in mirror are closer than they appear" mirror on a car, or magnifying, like a make-up mirror).

On preview, I like andrew cooke's "looking through" explanation. Through the Looking Glass, indeed!
posted by zsazsa at 6:50 AM on July 19, 2005

if you really look at a mirror, you'll see the dust on its surface, and maybe some smudges on the glass. that is in focus. when you look through the mirror, you see things further away, that may be blurred.

when you look at a photo, you'll see a picture that's in focus. you cannot look through a photo, unless it is very shiny (like a mirror) in which case you'll see reflections of things far away, that might be blurred.

if you compare looking at with looking at, everything is ok. if you compare looking through with looking through everything is ok. what is confusing is comparing looking at with looking through.
posted by andrew cooke at 6:54 AM on July 19, 2005

Well, first of all, if we're talking about the inside rearview mirror, I don't think the objects should appear any smaller. A planar mirror does not change the size of the object in question. It also neither focuses nor defocuses the rays, so your eye will do no better or worse at unblurring the image.
If you mean the sideview mirror, there's an added problem - they're actually convex, giving a smaller image and defocusing light. Assuming from your first statement that you're a bit nearsighted, this actually could make the objects a bit less blurry.

In a camera, on the other hand, there is definitely focusing going on. The light is focused with a converging lens onto a detector (film or electronic). Through various means, the picture is enlarged and printed, and you end up with your photograph. Which, when you look at it, is the same as looking at a drawing - you're not depending on the incident angle of light and in what direction it's reflected to form an image - you're looking just at the patterns of absorption on your photograph. You're looking at an object that is physically near, rather than an image.

On preview, andrew cooke beat me in pretty much every way.
posted by solotoro at 7:02 AM on July 19, 2005

The way I would explain it is... the light from a photograph starts AT the photograph (18 inches away perhaps) and ends at your eye (regardless of the whatever is in the photo). The light from a car, 6 or 7 lengths back, starts AT the car, 50-60 feet behind you and ends at your eye. The mirror does nothing more than redirect the light from going in direction 'A' to direction 'B', but the distance is stil "far". you can either turn your head around and look at the car directly or look at it in the mirror, but the light from the car is still traveling approximately the same distance.
posted by Necker at 10:50 AM on July 19, 2005

the light from whatever is in the photo started off there, and hit the camera. the camera recorded it and later it was printed as a photo. then light from the photo travels to the eye.

so why isn't the light from object to camera is important? what is it about the recording of the light in a photo that means we can ignore that? why is it wrong to think of the mirror as a magic film/camera/photo that both records and displays at the same time?

the difference comes down to a loss of phase information. light is waves - they have both an amplitude (how bright they are) and a phase (which is whether the "peaks" in one light wave are aligned with the peaks in another, or not, and by how much).

turns out that photographs don't record the phase information. that is enough to make eveything different, and frankly it blows my tiny mind.

something does record phase information - holograms. if you looked at a hologram of a distant place (impossible to generate using currently technology, i believe) then it would be blurred in the distance. i think - not 100% sure. this is a really good question!

this same argument underlies quantum mechanics and the tranition from "quantum" to "classical", iirc. if anyone is interested, think of/google feyman path integrals...
posted by andrew cooke at 12:05 PM on July 19, 2005

the difference comes down to a loss of phase information. ... photographs don't record the phase information.

It's true that photographs don't record the phase informatoin, but our eyes can't see that anyway. What makes photographs different is the lens and aperture on the camera.
posted by kindall at 2:16 PM on July 19, 2005

i see your point, but i think you miss mine.

at least, i think so. you seem to be saying that a camera focuses the image down to the focal plane. if you put the same lenses in front of a mirror then you would focus the scene onto the surface of the mirror.

that's true, but it means that if you looked through the mirror you'd see a blurry mess (looking at the mirror, necessary to see the focussed image, is hard, because the mirror scatters little light).

my point is that the blurry mess that you see when you look through the mirror is not what you see when you look through a photo. and the reason is phase information, i think, whether your eyes "can see it" or not.
posted by andrew cooke at 3:16 PM on July 19, 2005

another way of saying things - phase is not important at the focal plane, but the mirror is not at the focal plane when you look through it. you (kindall) are saying that (it's not at the focal plane) is the main difference (which is a reasonable point of view, i admit). i'm saying: "ok, but why can't a photo work like a mirror, too?" why isn't it possible to look through the photo and see the reflection of the lenses in front of the camera?
posted by andrew cooke at 3:26 PM on July 19, 2005

It is. That's called a hologram.
posted by kindall at 5:01 PM on July 19, 2005

But a photo isn't a hologram.

The point above that light doesn't have an odometer isn't exactly accurate. Light has an odometer, phase, that gets reset very frequently, once every wavelength.

andrew cooke's looking through vs. looking at could be made a little more formal by defining scenes and images (I have no idea if those are accepted terms...). The lenses in our eyes form an image from a scene, and that is what we 'see'. When we look at a mirror we are also looking at an image of a scene. When we look at a photograph we are looking at an image of an image - the cameras lens has already created the second image.

I'm sure it is true that phase is the key difference... I guess when you take an image you are throwing away the phase information that would naturally be part of the scene.

Anyway, I think we are still missing something...
posted by Chuckles at 6:49 PM on July 19, 2005