[AggieDad]

I was re-reading some articles in preparation for a project I am doing and came across this piece from Cambridge in Colour discussing the camera lens and the eye. I find it interesting and useful. Perhaps some of you will also.

CAMERAS VS. THE HUMAN EYE

Don

---

Last edited by AggieDad; 06-07-2019 at 07:20 AM.

[RGlasel]

I find it interesting and useful. Perhaps some of you will also.

Absolutely. This is a great article, thanks for posting the link. Some of this I already knew, but the illustrations and examples are really good at explaining things.

[normhead]

After just a quick reading will note few errors.

Counting the number of cones and rods etc, that actual number of "pixels" in the eye is about 7 MP, less than what can be displayed in a 4k monitor. The 52 MP estimation needs some clarification as to how they arrived at that number. It also might be noted in the section talking about the eye's ability to patch together different light values in a scene that our brain achieve what is in essence an internal HDR.

Another note...

The end result is a mental image whose detail has effectively been prioritized based on interest. This has an important but often overlooked implication for photographers: even if a photograph approaches the technical limits of camera detail, such detail ultimately won't count for much if the imagery itself isn't memorable.

This is the source of conflict between documentary and artistic photographer and "what they saw." The human brain emphasizes what it sees by filtering out uninteresting elements. Yet there is a certain school of thought that says when a photographer does that, producing an image "that emphasizes one part of the picture over another", they are accused of manipulating the scene. The basic thought there being that what the camera records is more valuable than what the human sees. Where as a more artistic photographer is going to argue, what he saw, is more valuable than all the extraneous information the camera recorded.

The thought of the day on this process, is that being a human, the photographer may not have seen everything was in the image, and may discover elements he/she didn't see when taking the photography that he then has to decide whether to keep emphasize or de-emphasize. The archival photographer will argue that the un-recognized element of the photo should be preserved even though he/she didn't "see" it. And in that case they are actually preserving detail for something they didn't see, not preserving what they saw.

What's funny to me is the question asked by the archivist types, does it look like what you saw? "Well know, I never saw that little kid picking his nose in the back ground". So if I didn't see him, is it OK if I photoshop him out? After all, that's not what I saw.
Or does "what you saw" only apply to some imaginary me that has a brain that doesn't filter what it sees.?

Especially relevant if you saw the beauty of a scene and didn't even see some of the more distracting elements it contained. I was trying to make an image of the beauty I saw. An element that would distract from that , wasn't seen and isn't important, because when I looked at the scene I didn't see it. When I took the image I didn't even see those tree branches in the top left corner of the sunset. So I take them out. I'm trying to capture my reality, not some kind of objective document of the scene that ignores my brains ability to filter and which was the original trigger of the endorphin release that helped me determine the image was worth capturing.

You want the viewer to experience what you experienced, unencumbered by extraneous elements your brain ignored.

Overall, most of the advantages of our visual system stem from the fact that our mind is able to intelligently interpret the information from our eyes, whereas with a camera, all we have is the raw image. Even so, current digital cameras fare surprisingly well, and surpass our own eyes for several visual capabilities. The real winner is the photographer who is able to intelligently assemble multiple camera images — thereby surpassing even our own mental image.

I can hear the howls of outrage.

The one corollary to this being the Human eye if memory serves me well doesn't have as much raw DR as a K-1 (if you take away it's ability to quickly adjust to high dynamic range situations and compose a composite image on the canvas in your brain), by a pretty significant margin. Modern high DR cameras provide much of the same "extra HDR" that you had to combine images to achieve in the past.

I haven't found an image that needed an HDR treatment since I bought my K-1.

[Apet-Sure]

Thanks for the link Don. I had forgotten about Cambridge in Colour. Just like opening an aperture lets in more light, articles like this open our understanding of what's going on when we take a picture. The whole process involves at least five elements: lens - sensor - camera processor - eye - brain. Each element has its own strengths and weaknesses. We juggle all this each time we snap the shutter.


As a side note, I find it fascinating to compare our human visual system to others in nature. Some birds of prey can actually change the focal length of their eyes and 'zoom in' on prey; helpful when you are 100 feet off the ground. We see red, green, blue, and black/white. As I understand it, birds can see UV light frequencies. Then there's the mantis shrimp. Marine biologists claim that they have 16 different color receptors compared to our three. How many trillions of colors can they differentiate, and why do they need to? Inquiring minds want to know.

[RGlasel]

Yet there is a certain school of thought that says when a photographer does that, producing an image "that emphasizes one part of the picture over another", they are accused of manipulating the scene. The basic thought there being that what the camera records is more valuable than what the human sees. Where as a more artistic photographer is going to argue, what he saw, is more valuable than all the extraneous information the camera recorded.

This is the crux of the matter, only someone whose frame of reference has been so severely constrained by a linear education, would fail to see that "manipulating the scene" happens every time the human brain decodes the signals from our our optic nerves (and there is no understanding without the involvement of our brains) and insisting "that what the camera records is more valuable than what the human sees" is a terrible form of blindness. As I see it, thinking in terms of "extraneous information" is just as bad, the real world is full of information we don't understand; to focus only on visual elements we can control blinds us to gaining understanding. Real art gives me a perspective that I can't see on my own, it doesn't dictate what I'm supposed to see.

The one thing I didn't find a mention of is that we have expectations of what a photograph should look like, which is very different from what we look for when we use own eyes and is very different between different people. Just because those of us who frequent this place tend to look for the same things in photographs, doesn't mean everyone else is also looking for those same things.

[normhead]

This is the crux of the matter, only someone whose frame of reference has been so severely constrained by a linear education, would fail to see that "manipulating the scene" happens every time the human brain decodes the signals from our our optic nerves (and there is no understanding without the involvement of our brains) and insisting "that what the camera records is more valuable than what the human sees" is a terrible form of blindness. As I see it, thinking in terms of "extraneous information" is just as bad, the real world is full of information we don't understand; to focus only on visual elements we can control blinds us to gaining understanding. Real art gives me a perspective that I can't see on my own, it doesn't dictate what I'm supposed to see.

The one thing I didn't find a mention of is that we have expectations of what a photograph should look like, which is very different from what we look for when we use own eyes and is very different between different people. Just because those of us who frequent this place tend to look for the same things in photographs, doesn't mean everyone else is also looking for those same things.

Another thought would be, as often happens when out with Tess, she'll say something like "what an awesome sunset." An emotional response. From a photographer's point of view , if the viewer doesn't look at the scene and say "what an awesome sunset", have you truly captured the scene? Is the goal to trigger a similar emotional response? Because if so, if you stop post processing before your image achieves that response, have you not done an emotional disservice to your viewer?

[timb64]

Another thought would be, as often happens when out with Tess, she'll say something like "what an awesome sunset." An emotional response. From a photographer's point of view , if the viewer doesn't look at the scene and say "what an awesome sunset", have you truly captured the scene? Is the goal to trigger a similar emotional response? Because if so, if you stop post processing before your image achieves that response, have you not done an emotional disservice to your viewer?

Sorry Norm,but it's not quite clear what you're saying there!Is the "photographer's" response to a beautiful sunset essentially any different to an "emotional"/non photographer's/Tess's response?Do we really compartmentalise our reactions in that way?

[normhead]

Sorry Norm,but it's not quite clear what you're saying there!Is the "photographer's" response to a beautiful sunset essentially any different to an "emotional"/non photographer's/Tess's response?Do we really compartmentalise our reactions in that way?

I didn't compartmentalize. Take another shot at trying to understand before I go at it from another way.

The photographers response tends to be a bit clipped as he/she reaches for the camera.

[AggieDad]

After just a quick reading will note few errors.
Counting the number of cones and rods etc, that actual number of "pixels" in the eye is about 7 MP, less than what can be displayed in a 4k monitor. The 52 MP estimation needs some clarification as to how they arrived at that number.

I am not sure you can (actually I am pretty sure you can't) equate rods and cones to pixels. Rods and cones are of different sizes and have totally different purposes within the eye.

And the article itself takes pains to explain that the 52 megapixel figure is not something for a direct comparison to a camera's pixel count:

"However, such calculations are misleading. Only our central vision is 20/20, so we never actually resolve that much detail in a single glance. Away from the center, our visual ability decreases dramatically, such that by just 20° off-center our eyes resolve only one-tenth as much detail.

Taking the above into account, a single glance by our eyes is therefore only capable of perceiving detail comparable to a 5-15 megapixel camera (depending on one's eyesight). However, our mind doesn't actually remember images pixel by pixel; it instead records memorable textures, color and contrast on an image by image basis."

So this might be a point of contention and even disagreement, but I certainly would not call it an error.

Don

[Zooland]

the equivalent daylight ISO for the human eye is even thought to be as low as 1.

I would love to see the day when camera sensors can match this.

[photoptimist]

Very roughly speaking, you could think of the human eye as three cameras merged into one:
* fovea camera: about a 1 MPix RGB sensor with a 300mm lens
* macula camera: about a 2 MPix monochrome+RGB camera with a 70mm lens
* peripheral camera: a 2 MPix monochrome camera with a 8mm fisheye lens

Only the tiniest central spot of vision sees high-resolution color. Human vision as a process basically does a Brenizer tiling of the scene -- moving the eye to center the high-res foveal spot on anything of interest. A large screen monitor or print must have many more MPix than the eye because it must support scrutiny by that fovea camera.

Also, at some level, the cones are like silicon pixels with each cone sending a signal to the brain and the rods are more like very sensitive B&W film grains with the signals of several rods aggregated to form a gray-level signal that is sent to the brain.

[normhead]

A large screen monitor or print must have many more MPix than the eye because it must support scrutiny by that fovea camera.

On that we differ. My point would be that the only meaningful viewing is from a "normal viewing distance" where the overall composition of the print is appreciated. Moving in to examine the print closely is not necessary. It seems to be insisted on by some photographers, but it's not at all necessary to the viewing public, including many of those who might want to purchase a print.

Resolution is that last thing to be considered in the purchase of a print, and the one most easily ignored.

WHo took the world's most expensive photograph? - PentaxForums.com

High resolution is really only necessary if you want 3 million for your print (Gursky) instead of 1 million,(Peter Lik). Given the cost of the equipment (not to mention the secret post processing process) needed to get 3 million, I think 1 million will do for most of us.

[photoptimist]

On that we differ. My point would be that the only meaningful viewing is from a "normal viewing distance" where the overall composition of the print is appreciated. Moving in to examine the print closely is not necessary. It seems to be insisted on by some photographers, but it's not at all necessary to the viewing public, including many of those who might want to purchase a print.

Resolution is that last thing to be considered in the purchase of a print, and the one most easily ignored.

WHo took the world's most expensive photograph? - PentaxForums.com

High resolution is really only necessary if you want 3 million for your print (Gursky) instead of 1 million,(Peter Lik). Given the cost of the equipment (not to mention the secret post processing process) needed to get 3 million, I think 1 million will do for most of us.

Every bit of what I said applies to "normal viewing distances." If you have the equivalent of 300 mm lens on a 1 MPix full-frame camera held at "normal viewing distances," you'll see that the resolution on the screen or print surface needs to be on the order of 100 to 200 PPI or the eye will see some blur or fuzziness. The human eye, when it looks at a scene, a print, or a monitor is essentially tiling together literally dozens of 1 MPix Brenizer & focus-stacked frames with the brain building up a very high MPix version of the scene inside your head.

Where we agree about resolution is that not all prints or computer images have (or need) that high-MPix resolution -- softness or blur is a virtue in some types of images. But unless one knows that 100% of the images being printed or displayed are of the soft-and-blurry kind, the printer or monitor needs to support a sufficiently high resolution to do a good job on the high resolution scenes.

Resolution is also a personal choice -- I absolutely love high-resolution images. Resolution really is among my top criteria when I consider a print of a landscape, cityscape, macro, or astro scene.

[beholder3]

A large screen monitor or print must have many more MPix than the eye because it must support scrutiny by that fovea camera.

Yes to the need of a picture to allow full eye resolution.

But no in the context of the linked article.

Even the central vision capabilities of a human are quite limited (in the range of something that means you will not overresolve details on a 6-7 MPx image even when your eyes move around it).
It's all simple math.

To test it and to base your own calculations you can simply test it by trying to actually count (not just notice the group of markings) the millimeter markings on a ruler from some distance in good light.
With normal visual acuity (the 6-7 Mpx one) you are expected to be able to resolve the markings from somewhere around 150 cm distance.
Obviously this only is a ballpark estimate then, but it is easy to do.

[normhead]

Every bit of what I said applies to "normal viewing distances." If you have the equivalent of 300 mm lens on a 1 MPix full-frame camera held at "normal viewing distances," you'll see that the resolution on the screen or print surface needs to be on the order of 100 to 200 PPI or the eye will see some blur or fuzziness. The human eye, when it looks at a scene, a print, or a monitor is essentially tiling together literally dozens of 1 MPix Brenizer & focus-stacked frames with the brain building up a very high MPix version of the scene inside your head.

Where we agree about resolution is that not all prints or computer images have (or need) that high-MPix resolution -- softness or blur is a virtue in some types of images. But unless one knows that 100% of the images being printed or displayed are of the soft-and-blurry kind, the printer or monitor needs to support a sufficiently high resolution to do a good job on the high resolution scenes.

Resolution is also a personal choice -- I absolutely love high-resolution images. Resolution really is among my top criteria when I consider a print of a landscape, cityscape, macro, or astro scene.

You lost me, do you have a real world example?

The human eye, when it looks at a scene, a print, or a monitor is essentially tiling together literally dozens of 1 MPix Brenizer & focus-stacked frames with the brain building up a very high MPix version of the scene inside your head.

Mine would be a 540 x 380 image I took with an old digital camera that I blew up to a 4x6. Even at 18 inches held in my hand it didn't appear to be soft or fuzzy. That was printed at 72 DPI. And it was hanging on our wall for almost 5 years. I just have no real world experience that would let me agree with your post.

The human eye, when it looks at a scene, a print, or a monitor is essentially tiling together literally dozens of 1 MPix Brenizer & focus-stacked frames with the brain building up a very high MPix version of the scene inside your head.

Where's there research on this? That's a pretty involved theory to be tossing out there without empirical confirmation. I'm not disputing that how the brain works in complaining images, that's pretty well understood. What I'm disputing is that the image formed in the brain is augmented by oversampling the image as opposed to lower res. In fact I'd argue , if anything, the brain can only process what the eye can detect, and that detail too small to be detected cannot possibly affect the final "brain image".

It needs to be acknowledged that there can be too much resolution and that resolution beyond what the brain can resolve at a normal viewing distance is wasted. You seem to be implying that the brain can piece together information the eye cannot resolve.

Unlike yourself, I've been remarkably un-impressed with the difference between my K-5, K-3 and K-1 prints. I'm still happy with K-5 prints, and many of our prints are from 12 MP cameras. I see the progress toward higher res a due to the marketing genius of camera companies and something that very rarely affects the impact of a print it in fact, it ever does. We passed the necessary resolution for pleasing 30x20 prints when we passed 12 MP, and possible after 10 MP.