Dynamic Range of the Human Eye

Started Jan 16, 2018 | Polls
Hulk2006
Hulk2006 Regular Member • Posts: 103
Dynamic Range of the Human Eye
2

I've always wondered how many stops a digital camera would need to have in order to have the same dynamic range as the human eye.  What do you guys think? Do you think we will ever get there and maybe even go beyond?

POLL
12-14 f-stops
15.6% 12  votes
15-17 f-stops
13.0% 10  votes
18-20 f-stops
26.0% 20  votes
21-23 f-stops
20.8% 16  votes
24-26 f-stops
3.9% 3  votes
27+ f-stops
20.8% 16  votes
  Show results
J A C S
J A C S Forum Pro • Posts: 18,178
Re: Dynamic Range of the Human Eye
23

None of the above. 6.5 stops. But we can see much more by scanning the scene.

Read more here:

https://en.wikipedia.org/wiki/Human_eye#Dynamic_range

Lee Jay Forum Pro • Posts: 56,673
Re: Dynamic Range of the Human Eye
1

J A C S wrote:

None of the above. 6.5 stops. But we can see much more by scanning the scene.

Read more here:

https://en.wikipedia.org/wiki/Human_eye#Dynamic_range

I read so much different stuff about this, I decided to test it myself. For example:

https://wolfcrow.com/blog/notes-by-dr-optoglass-dynamic-range-of-the-human-eye/

"This means that the human eye has a greater dynamic range in the shadows than in bright light – but only when no bright light is directly hitting our retina – the dynamic range of the eye in the shadows is about 20 stops. Which means the dynamic range of the eye in bright outdoor sunlight is only about 10 stops."

My own tests of my own eye, in middle light, one eye covered, staring with the other one, was about 14 stops.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon Extender EF 2x III Canon EF 8-15mm f/4L Fisheye USM +10 more
J A C S
J A C S Forum Pro • Posts: 18,178
Re: Dynamic Range of the Human Eye

The answer depends on how you define it. The human eye can adapt to different light levels, can scan the scene, is affected by flare, etc.

dwalby Veteran Member • Posts: 5,895
Re: Dynamic Range of the Human Eye
2

J A C S wrote:

The answer depends on how you define it. The human eye can adapt to different light levels, can scan the scene, is affected by flare, etc.

the big difference is the brain translates things for the eye that it really doesn't see, based on experience.  The classic example of this is the cube with the same shades in the shadows and in the direct light, the eye/brain interprets those identical shades as being dramatically different.  That probably has at least as much, if not more, impact on the 'dynamic range' of the human eye as the eye itself.

OpticsEngineer Veteran Member • Posts: 7,013
Re: Dynamic Range of the Human Eye
4

My personal observations are more in line with yours. The 100:1 in the Wikipedia article is clearly wrong although I have seen it cited many times over the years. But the meaning has been quite different that what the OP asked. The100:1 goes back to black and white TV work concerning how many distinct grey levels were needed to create a realistic appearing scene in video. And that is all you will get from an old black and white TV. But the way those old journal articles were written it is easy to misinterpret them if you don't understand the goal of the research... the context is left out of the articles.

A more recent work more in line with the OPs question would 5000:1 in a single scene.

"The Dynamic Range of Human Lightness Perception"  Ana Radonjić,1 Sarah R. Allred,2 Alan L. Gilchrist,3 and David H. Brainard1

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3244211/

You can skip down to the Highlights section and read:

"The dynamic range of lightness perception exceeds 5,000:1 in a single image context"

As a simple experiment, to demonstrate the eye can handle a lot more dynamic range than 100:1, I would encourage one to try this. Look out a window on a normal sunny day. Notice you can see detail in the rim of the window at the same time you can see detail in the scene outdoors. Even better if you have a power meter handy with something like a 1 degree field of view, measure the luminance on the window sill rim and in the outdoor scene. I used a Minolta LS-100 when I did it a few years ago. I don't recall what value I got but I do recall it was much much more than 100:1.

 OpticsEngineer's gear list:OpticsEngineer's gear list
Olympus XZ-2 iHS Fujifilm XF1 Canon PowerShot G7 X Olympus Tough TG-4 Sony SLT-A65 +27 more
SpiltMilk Senior Member • Posts: 1,644
Re: Dynamic Range of the Human Eye
1

dwalby wrote:

J A C S wrote:

The answer depends on how you define it. The human eye can adapt to different light levels, can scan the scene, is affected by flare, etc.

the big difference is the brain translates things for the eye that it really doesn't see, based on experience. The classic example of this is the cube with the same shades in the shadows and in the direct light, the eye/brain interprets those identical shades as being dramatically different. That probably has at least as much, if not more, impact on the 'dynamic range' of the human eye as the eye itself.

Thia?

Yes - it's almost impossible to define properly as subjective perception takes over.

What's the DR of that image?  Two widely different shades there, top panel and bottom panel.

Then  you put your finger over the middle join piece and... lo and behold!  Same shade.

-- hide signature --

Your first mistake was thinking that I'm one of your sheep...

Fotoni Senior Member • Posts: 1,005
Re: Dynamic Range of the Human Eye

I have used five bracketed photos which start to look quite same as how I see when shooting against the sun. D3200 has about 13 stops of dynamic range according to the dxomark.com and five brackets give extra 8 stops, so I would say that human vision has +20 stops of dynamic range.

 Fotoni's gear list:Fotoni's gear list
Pentax K-1 Pentax FA* 50mm F1.4 SDM AW Canon PowerShot SD4000 IS
Lee Jay Forum Pro • Posts: 56,673
Re: Dynamic Range of the Human Eye

J A C S wrote:

The answer depends on how you define it. The human eye can adapt to different light levels, can scan the scene, is affected by flare, etc.

True, but I'm talking about one scene with one eye at one time without scanning.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon Extender EF 2x III Canon EF 8-15mm f/4L Fisheye USM +10 more
J A C S
J A C S Forum Pro • Posts: 18,178
Re: Dynamic Range of the Human Eye

OpticsEngineer wrote:

My personal observations are more in line with yours. The 100:1 in the Wikipedia article is clearly wrong although I have seen it cited many times over the years. But the meaning has been quite different that what the OP asked. The100:1 goes back to black and white TV work concerning how many distinct grey levels were needed to create a realistic appearing scene in video. And that is all you will get from an old black and white TV. But the way those old journal articles were written it is easy to misinterpret them if you don't understand the goal of the research... the context is left out of the articles.

A more recent work more in line with the OPs question would 5000:1 in a single scene.

"The Dynamic Range of Human Lightness Perception" Ana Radonjić,1 Sarah R. Allred,2 Alan L. Gilchrist,3 and David H. Brainard1

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3244211/

You can skip down to the Highlights section and read:

"The dynamic range of lightness perception exceeds 5,000:1 in a single image context"

As a simple experiment, to demonstrate the eye can handle a lot more dynamic range than 100:1, I would encourage one to try this. Look out a window on a normal sunny day. Notice you can see detail in the rim of the window at the same time you can see detail in the scene outdoors. Even better if you have a power meter handy with something like a 1 degree field of view, measure the luminance on the window sill rim and in the outdoor scene. I used a Minolta LS-100 when I did it a few years ago. I don't recall what value I got but I do recall it was much much more than 100:1.

"One can handle" and "lightness perception" is a different than what Wikipedia page says - they talk about DR of the retina. I found a reference here to this book: The First Steps in Seeing by R. W. Rodieck, where they claim 100:1. So it is not "obviously wrong".

As I said, this is a loaded question. The answer depends on the way it is asked. "DR of the eye" has no much meaning by itself.

Lee Jay Forum Pro • Posts: 56,673
Re: Dynamic Range of the Human Eye

J A C S wrote:

OpticsEngineer wrote:

My personal observations are more in line with yours. The 100:1 in the Wikipedia article is clearly wrong although I have seen it cited many times over the years. But the meaning has been quite different that what the OP asked. The100:1 goes back to black and white TV work concerning how many distinct grey levels were needed to create a realistic appearing scene in video. And that is all you will get from an old black and white TV. But the way those old journal articles were written it is easy to misinterpret them if you don't understand the goal of the research... the context is left out of the articles.

A more recent work more in line with the OPs question would 5000:1 in a single scene.

"The Dynamic Range of Human Lightness Perception" Ana Radonjić,1 Sarah R. Allred,2 Alan L. Gilchrist,3 and David H. Brainard1

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3244211/

You can skip down to the Highlights section and read:

"The dynamic range of lightness perception exceeds 5,000:1 in a single image context"

As a simple experiment, to demonstrate the eye can handle a lot more dynamic range than 100:1, I would encourage one to try this. Look out a window on a normal sunny day. Notice you can see detail in the rim of the window at the same time you can see detail in the scene outdoors. Even better if you have a power meter handy with something like a 1 degree field of view, measure the luminance on the window sill rim and in the outdoor scene. I used a Minolta LS-100 when I did it a few years ago. I don't recall what value I got but I do recall it was much much more than 100:1.

"One can handle" and "lightness perception" is a different than what Wikipedia page says - they talk about DR of the retina. I found a reference here to this book: The First Steps in Seeing by R. W. Rodieck, where they claim 100:1. So it is not "obviously wrong".

As I said, this is a loaded question. The answer depends on the way it is asked. "DR of the eye" has no much meaning by itself.

Maybe they're talking about a single cone.  However, that's largely irrelevant because different parts of the retina can run at different sensitivity levels simultaneously.  Even in very bright light (ranging from direct sun to indoor) I can see about 14 stops of DR only one degree apart.  I haven't tested it in dark conditions.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon Extender EF 2x III Canon EF 8-15mm f/4L Fisheye USM +10 more
J A C S
J A C S Forum Pro • Posts: 18,178
Re: Dynamic Range of the Human Eye

Lee Jay wrote:

J A C S wrote:

OpticsEngineer wrote:

My personal observations are more in line with yours. The 100:1 in the Wikipedia article is clearly wrong although I have seen it cited many times over the years. But the meaning has been quite different that what the OP asked. The100:1 goes back to black and white TV work concerning how many distinct grey levels were needed to create a realistic appearing scene in video. And that is all you will get from an old black and white TV. But the way those old journal articles were written it is easy to misinterpret them if you don't understand the goal of the research... the context is left out of the articles.

A more recent work more in line with the OPs question would 5000:1 in a single scene.

"The Dynamic Range of Human Lightness Perception" Ana Radonjić,1 Sarah R. Allred,2 Alan L. Gilchrist,3 and David H. Brainard1

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3244211/

You can skip down to the Highlights section and read:

"The dynamic range of lightness perception exceeds 5,000:1 in a single image context"

As a simple experiment, to demonstrate the eye can handle a lot more dynamic range than 100:1, I would encourage one to try this. Look out a window on a normal sunny day. Notice you can see detail in the rim of the window at the same time you can see detail in the scene outdoors. Even better if you have a power meter handy with something like a 1 degree field of view, measure the luminance on the window sill rim and in the outdoor scene. I used a Minolta LS-100 when I did it a few years ago. I don't recall what value I got but I do recall it was much much more than 100:1.

"One can handle" and "lightness perception" is a different than what Wikipedia page says - they talk about DR of the retina. I found a reference here to this book: The First Steps in Seeing by R. W. Rodieck, where they claim 100:1. So it is not "obviously wrong".

As I said, this is a loaded question. The answer depends on the way it is asked. "DR of the eye" has no much meaning by itself.

Maybe they're talking about a single cone.

It is in the first link.

However, that's largely irrelevant because different parts of the retina can run at different sensitivity levels simultaneously. Even in very bright light (ranging from direct sun to indoor) I can see about 14 stops of DR only one degree apart. I haven't tested it in dark conditions.

So it depends on how the question is asked? BTW, just because you can see details in a 14 stops scene, if you can, does not mean that you have 14 stops DR. The darks are polluted by a lot of flare. What is projected on your retina is much less than 14 stops. Also, if there is a very high local contrast, your DR would be reduced, like on the screen you are looking at now.

tbcass
tbcass Forum Pro • Posts: 48,902
Not as much as you think.

The dynamic range of the human eye isn't as much as people think. It only seems like it's greater because your iris opens and closes automatically giving the false impression of more much like a video camera constantly adjusts for lighting changes. Your digital still camera does the same but of course the moment of photo capture requires it to stop adjusting. If you combine sensor DR, shutter speed, and f stop the DR of the camera completely blows the eyes capabilities away and is way more than any of your choices.

-- hide signature --

Tom
Look at the picture, not the pixels

 tbcass's gear list:tbcass's gear list
Sony RX100 Sony RX10 IV Sony a99 II Tamron SP AF 90mm F/2.8 Di Macro Tamron SP 70-300mm F/4-5.6 Di USD +8 more
Lee Jay Forum Pro • Posts: 56,673
Re: Dynamic Range of the Human Eye
3

I have a window sill right next to me and a north-facing window.  Right now I can put a magazine with a black background in the shadow under the sill and read the text clearly while at the same moment see the tree outside and the blue (not white) sky.

Magazine exposure: 1/4, ISO 800, f/5.6

Outside exposure: 1/320, ISO 100, f/5.6

That's over 9 stops right there, and I can see highlights outside and shadows on the magazine at the same instant with the same eye without scanning.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon Extender EF 2x III Canon EF 8-15mm f/4L Fisheye USM +10 more
tbcass
tbcass Forum Pro • Posts: 48,902
Re: Dynamic Range of the Human Eye

10-14 stops seems to be the most accepted range for the human eye.

-- hide signature --

Tom
Look at the picture, not the pixels

 tbcass's gear list:tbcass's gear list
Sony RX100 Sony RX10 IV Sony a99 II Tamron SP AF 90mm F/2.8 Di Macro Tamron SP 70-300mm F/4-5.6 Di USD +8 more
Kotagrapher18 New Member • Posts: 3
variable/dynamic iris

tbcass wrote:

The dynamic range of the human eye isn't as much as people think. It only seems like it's greater because your iris opens and closes automatically giving the false impression of more much like a video camera constantly adjusts for lighting changes. Your digital still camera does the same but of course the moment of photo capture requires it to stop adjusting. If you combine sensor DR, shutter speed, and f stop the DR of the camera completely blows the eyes capabilities away and is way more than any of your choices.

Eyes have variable/dynamic irises.   They change "aperture" quickly to act like bracketing -5EV to +5EV images and combining to get an HDR JPEG.

tbcass
tbcass Forum Pro • Posts: 48,902
Re: variable/dynamic iris

Kotagrapher18 wrote:

tbcass wrote:

The dynamic range of the human eye isn't as much as people think. It only seems like it's greater because your iris opens and closes automatically giving the false impression of more much like a video camera constantly adjusts for lighting changes. Your digital still camera does the same but of course the moment of photo capture requires it to stop adjusting. If you combine sensor DR, shutter speed, and f stop the DR of the camera completely blows the eyes capabilities away and is way more than any of your choices.

Eyes have variable/dynamic irises. They change "aperture" quickly to act like bracketing -5EV to +5EV images and combining to get an HDR JPEG.

Not true because it's dynamic, not a single still image with your eye which behaves more like a video camera. At any given moment in time the eye DR is only 10-14 ev.

-- hide signature --

Tom
Look at the picture, not the pixels

 tbcass's gear list:tbcass's gear list
Sony RX100 Sony RX10 IV Sony a99 II Tamron SP AF 90mm F/2.8 Di Macro Tamron SP 70-300mm F/4-5.6 Di USD +8 more
J A C S
J A C S Forum Pro • Posts: 18,178
Re: Dynamic Range of the Human Eye
2

Lee Jay wrote:

I have a window sill right next to me and a north-facing window. Right now I can put a magazine with a black background in the shadow under the sill and read the text clearly while at the same moment see the tree outside and the blue (not white) sky.

Magazine exposure: 1/4, ISO 800, f/5.6

Outside exposure: 1/320, ISO 100, f/5.6

That's over 9 stops right there, and I can see highlights outside and shadows on the magazine at the same instant with the same eye without scanning.

So all those researchers are wrong?

EDIT: Forget about 14 stops, let us try 6 or so. Can you see the pattern here? I made it so that you can barely see but you still can see the one in the horizontal bar.

How about here? The "H" is the same; just the background is different.

Lee Jay Forum Pro • Posts: 56,673
Re: Dynamic Range of the Human Eye
1

J A C S wrote:

Lee Jay wrote:

I have a window sill right next to me and a north-facing window. Right now I can put a magazine with a black background in the shadow under the sill and read the text clearly while at the same moment see the tree outside and the blue (not white) sky.

Magazine exposure: 1/4, ISO 800, f/5.6

Outside exposure: 1/320, ISO 100, f/5.6

That's over 9 stops right there, and I can see highlights outside and shadows on the magazine at the same instant with the same eye without scanning.

So all those researchers are wrong?

Every bit of research I can find on the subject gives the dynamic range of human vision as 10 stops or more.  And I've found a lot.  I measured 9 above just from middle gray to middle gray so my measurements are consistent with the research I've found.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon Extender EF 2x III Canon EF 8-15mm f/4L Fisheye USM +10 more
Lee Jay Forum Pro • Posts: 56,673
Re: Dynamic Range of the Human Eye

J A C S wrote:

Lee Jay wrote:

I have a window sill right next to me and a north-facing window. Right now I can put a magazine with a black background in the shadow under the sill and read the text clearly while at the same moment see the tree outside and the blue (not white) sky.

Magazine exposure: 1/4, ISO 800, f/5.6

Outside exposure: 1/320, ISO 100, f/5.6

That's over 9 stops right there, and I can see highlights outside and shadows on the magazine at the same instant with the same eye without scanning.

So all those researchers are wrong?

https://pdfs.semanticscholar.org/50bf/f7d5f4d8f3f230b6f446e667dd22f12567dc.pdf

"The eye can capture approximately 5 orders of magnitude of dynamic range effectively simultaneously."

That's 16 2/3 stops.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon Extender EF 2x III Canon EF 8-15mm f/4L Fisheye USM +10 more
Keyboard shortcuts:
FForum MMy threads