Human eyes: 576 megapixels

[Werra]

I don't really understand how this figure is arrived at but it seems an interesting factoid to me!

"How many megapixels do our eyes have?"
Sound and Light Reflections: How many megapixels do our eyes have?

• comparing the number of sensing elements - the eye has 5 million cones (the color receptors) and 100 million rods (the monochrome contrast receptors) which give a human the equivalent of two 105MP (MegaPixel) video cameras (because we have two eyes).
• comparing spatial resolution over the eye's field of view - we have the equivalent of 576MP.

 

[Garbz]

Clarkvision Photography - Resolution of the Human Eye This article mentions how they come by that figure. Which may I say is completely and utterly crap.

The equivalent is saying that my camera has near infinite resolution since I can attach a 1200mm lens to it and make a 360degree panorama. The eye has a finite resolution, the fact that through movement it infers more data is irrelevant. We don't call a 1080p movie camera a 100mpxl camera simply because *gasp* we can move it around to see more.

Furthermore the comparison is ******** since it assumes that you have the same visual acuity over the visual range. Sure my eye can see about 180 degrees, but good luck reading a book that far out of the corner of your eye without turning to look at it. Outside our focus point our resolution is horrendous. The equivalent in camera terms would be two 105mpxl cameras with absolutely crap colour performance mounted next to each other each with a fisheye lens on it that look sharp only in the dead centre and around the edges may have the quality of a handmade piece of plastic.

 

[KmH]

I'm with Garbz.

That article is a prime example of

If you can't convince them with facts, dazzle 'em with BS.

"Human eyes: 576 megapixels" is total BS.

 

[Werra]

Maybe so but still more constructive than your input.

 

[analog.universe]

Pseudo-science fail


This reminds me of one those extra credit problems from high school physics: "Compute the energy output of Santa's reindeer necessary to deliver to every home" or "Determine whether hell is endothermic or exothermic"

It's just a way to have some fun with math, and has little or no connection to reality....

 

[Werra]

Thanks for the link. I think he puts forward an interesting idea. It isn't invalidated by what you could do with your camera and a 1200mm to make a panorama. I also think that's a very good analogy you make too. But the aggression and sarcasm in your responce isn't necessary. As far as i know, the distribution of rods and cones is an even pattern across the eye extending beyond a singular area of focus. Acuity isn't just a question of equivalent megapixels, neither is color rendition entirely relevant to both rods and cones. Judging your counter-arguments in the same way you judged Clarksons, I would rate yours as crappy too and certainly more partisan and negative.

cheers

Clarkvision Photography - Resolution of the Human Eye This article mentions how they come by that figure. Which may I say is completely and utterly crap.

The equivalent is saying that my camera has near infinite resolution since I can attach a 1200mm lens to it and make a 360degree panorama. The eye has a finite resolution, the fact that through movement it infers more data is irrelevant. We don't call a 1080p movie camera a 100mpxl camera simply because *gasp* we can move it around to see more.

Furthermore the comparison is ******** since it assumes that you have the same visual acuity over the visual range. Sure my eye can see about 180 degrees, but good luck reading a book that far out of the corner of your eye without turning to look at it. Outside our focus point our resolution is horrendous. The equivalent in camera terms would be two 105mpxl cameras with absolutely crap colour performance mounted next to each other each with a fisheye lens on it that look sharp only in the dead centre and around the edges may have the quality of a handmade piece of plastic.

 

[Garbz]

I was going for more sarcasm than aggression, but my intention really was to come across as negative since there really is little positive to say. Articles like this are slowly getting annoying, and the internet is unfortunately full of lots of opinions that look great when you add some maths despite the fact that the maths itself may be completely relevant.

But if the idea isn't invalidated by putting a 1200mm on my camera to make a panorama does that also mean I increase the "megapixel" value of my eyes by putting binoculars on? The whole problem I have with this is that we are trying to compare two completely different parts of a system. By saying the eyes have x number of megapixels because of how we look around and our field of view is like saying owning a Ford is better than owning a V6. Means absolutely nothing. You are comparing a system along with optics + flat out cheating with the whole ability to look around, to a pixel number on a sensor?

I liked the original analogy better which was the number of rods and cones in the eye but even that one doesn't hold very well in terms of photography (which I imagine is what the original question was going for) because of the distribution, which is definitely not even across the retina. Not just the density changes but the relative rod to cone ratio does too. The bit in the eye responsible for our accurate vision where we are focusing is called the fovea centralis and there are about 200x more cones in this region than the rest of the eye. Also it's not just more cones, but the cones are physically a very different shape too.

I hope you didn't take my comparison of the eye to a fisheye that is horrendously bad everywhere except in the dead centre sarcastically because I meant every word of that. It really acts like that... actually worse since the concentration of cones in the fovea also means that the middle of our eye is not only able to achieve a very different quality visually, but also able to resolve colour far better than our peripheral vision.


By the way trivia answer: now you should realise where the Foveon sensor got its name from, the part of the eye densely packed with pixels that can resolve full colour

 

[JG_Coleman]

I don't really understand how this figure is arrived at but it seems an interesting factoid to me!

"How many megapixels do our eyes have?"
Sound and Light Reflections: How many megapixels do our eyes have?

• comparing the number of sensing elements - the eye has 5 million cones (the color receptors) and 100 million rods (the monochrome contrast receptors) which give a human the equivalent of two 105MP (MegaPixel) video cameras (because we have two eyes).
• comparing spatial resolution over the eye's field of view - we have the equivalent of 576MP.

I remember reading an article similar to this some time ago. Admittedly, I was interested enough to read what was offered as a supporting argument.

Throughout sort of an on-going, informal comparison of how my camera sees the world as opposed to my eyes, though, I've come to the conclusion that the two are very difficult to compare... partially for reasons mentioned here already, but partially for other reasons that I've stumbled upon through attempting to "observe myself observing" (as convoluted as that is).

The biggest problem I've noticed in trying to come to some kind of comparative figure between cameras and human eyes is that human eyes, more often than not, "see" only what we desire that they see. That is, I realize that when I'm looking at something that has captured my attention, it tends to be the only thing in my field of vision that is properly resolved. I'm not sure that the human eye is necessarily this limited, in a mechanical sense (I have 20/20 vision, after all), but consciousness can actually filter out and ignore our periphery based upon what we are looking at. And, it would seem, our effective periphery can swell or shrink depending upon how intensely focused we are upon a given a thing and how far away that given thing may be.

There's really no way to compare a camera, which is only "selective" in the sense that it is limited by focal lengths and field-of-view and sensor receptors, to a human visual faculties which are extremely selective in what they receive and resolve based upon our intentions and interest on different cues in our external environment.

 

[KmH]

As far as i know, the distribution of rods and cones is an even pattern across the eye

Nope!.

Astronomers discovered long ago to use 'averted vision' to see faint objects in their telescopes.

The cones (color receptors) are concentrated in the middle of the retina (the fovea), and there are 3 distinct types. The rods (B&W receptors) cover the rest of the retina.

The cones don't work at all in low light, which is why we don't see colors at night, unless there is a colored light source.

Eye - Wikipedia, the free encyclopedia