How Does Reflectance Affect Exposure?

[andytakeone]

I know this is somewhat of a noobie question that I can find if I searched for it, but the things I've been reading recently have kind of been confusing me on this subject.

So I thought I might clear it up since you guys are good at being candid.

How does reflectance affect exposure? Do you only expose for incident light? Or does reflectance have a big impact on exposure? How big?

Thanks.

 

[tirediron]

I think you're departing from photography and moving into physics... just to clarify: Do you mean reflectance as in the highlights in the scene, or the actual reflectance of each different object/texture/tone?

 

[Big Mike]

Reflectance is a property of objects (things).

When light hits something; it can be absorbed, it can pass though (transmission) or it can reflect. Usually, it's some combination of the three.

Usually we are concerned with the reflection aspect, as that is what we see/photograph.

There is diffuse reflection & direct reflection, but lets leave that for another day.

Objects that exhibit more reflection, will look brighter....objects with less will look darker. I think that what you are asking is: how does having more or less reflection affect the exposure that we set on our cameras?

The simple answer is....it shouldn't affect your exposure.

To get a proper exposure (bright things looking bright, dark things looking dark) you should set your exposure based on the incident light....not the reflected light.

Now, with that being said...the most common way to figure out exposure is by using the reflected light meter built into your camera. So the task then becomes, figuring out how to use your reflected light meter to measure and expose for the incident light.

 

[dennybeall]

The camera is using ALL light that reaches it's sensor, incident or reflected. The reflectance of the object determines how much reflected light there is but the sensor on the camera just uses what comes to it.
Personally I rarely use an incident light meter anymore. The spot setting on the camera meter works 99% of the time.

 

[Ysarex]

I made the above illustration one afternoon for my class -- using a reflected light meter built into a camera. There are three photos there; The top scene shows two sedans parked across the street from my house. Below the silver sedan is a section of the door panel directly below the door handle. Below the charcoal sedan is a section of the door panel directly below the door handle.

First I took the photo of the entire scene including both cars using the camera meter's exposure determination. One car reflects more light than the other. With both cars in the scene the meter averages the reflectance of both. Then I walked across the street and got real close and photographed the silver car door panel -- then the charcoal car door panel. In these two photos I again used the camera meter's exposure determination. ALL THREE EXPOSURES WERE DIFFERENT. The exposure for the silver car door panel was less than the overall scene and the exposure for the charcoal door panel was more than the overall scene. The camera meter darkened the silver car's door panel and lightened the charcoal car's door panel.

Open the photo in an app like PS or GIMP where you can measure HSB (hue, saturation, brightness) values and check those four spots. In the top photo I get a brightness value for the silver car door: 53 and for the charcoal car door: 16 -- the point: that's a substantial difference as it should be -- the silver car reflects more light and is brighter. In the two closeups of the door panels I get silver car door: 38 and charcoal car door: 39.

Joe

 

[soufiej]

I know this is somewhat of a noobie question that I can find if I searched for it, but the things I've been reading recently have kind of been confusing me on this subject.

So I thought I might clear it up since you guys are good at being candid.

How does reflectance affect exposure? Do you only expose for incident light? Or does reflectance have a big impact on exposure? How big?

Thanks.

Let's take the image of the two cars in the above post to think about "reflectance". Exposure aside, what we see with the light colored car is greater reflectance as seen by the light meter reading. It exists at the top end of the dynamic range we may capture with our digital camera.

What we need to use is not so much the exposure value but the dynamic range of the subject, which is actually about average/equal for both cars.

Rather like a musical passage which begins at the bottom of the DR, a whisper , and a musical passage which exists at the very top of the DR, a shout.

If our equipment is not capable of the full DR, then we will run into problems. If the reflectance occurs at the top of the DR, we risk blowing out the highlights. If we are trying to push the limits of our equipment with the very lowest DR, then we risk losing detail in the shadows.

IMO, it is not the "exposure" of any image that really matters, it is the total dynamic range. Reflectance is a function of DR, not exposure. Exposure is flexible. Dynamic range less so.

Use your histogram. It will indicate when DR is greater than your camera can process.

Otherwise, every camera has its own DR limit. Two cameras using the equivalent "exposure" values - which can vary considerably depending on the photographer's intent for the image - can have very different DR limits.

This would mean the two cameras are reacting to reflectance in totally discrete ways. The histogram tells you when you have exceeded the DR your specific camera can deal with.

 

[Big Mike]

I made the above illustration one afternoon for my class -- using a reflected light meter built into a camera. There are three photos there; The top scene shows two sedans parked across the street from my house. Below the silver sedan is a section of the door panel directly below the door handle. Below the charcoal sedan is a section of the door panel directly below the door handle.

First I took the photo of the entire scene including both cars using the camera meter's exposure determination. One car reflects more light than the other. With both cars in the scene the meter averages the reflectance of both. Then I walked across the street and got real close and photographed the silver car door panel -- then the charcoal car door panel. In these two photos I again used the camera meter's exposure determination. ALL THREE EXPOSURES WERE DIFFERENT. The exposure for the silver car door panel was less than the overall scene and the exposure for the charcoal door panel was more than the overall scene. The camera meter darkened the silver car's door panel and lightened the charcoal car's door panel.

Open the photo in an app like PS or GIMP where you can measure HSB (hue, saturation, brightness) values and check those four spots. In the top photo I get a brightness value for the silver car door: 53 and for the charcoal car door: 16 -- the point: that's a substantial difference as it should be -- the silver car reflects more light and is brighter. In the two closeups of the door panels I get silver car door: 38 and charcoal car door: 39.

Joe

Great example Joe.

I would point to this as proof that we shouldn't rely on auto metering alone (which basically reacts to the reflectance of the scene/object) to determine exposure.

For more accurate metering, we either need to constantly think about how the camera's meter is interpreting the reflectance, and compensate for it....OR we need to use something (could be the in-camera meter) to determine the amount of incident light and expose for that (best done in manual mode).

 

[Ornello]

I know this is somewhat of a noobie question that I can find if I searched for it, but the things I've been reading recently have kind of been confusing me on this subject.

So I thought I might clear it up since you guys are good at being candid.

How does reflectance affect exposure? Do you only expose for incident light? Or does reflectance have a big impact on exposure? How big?

Thanks.

It really doesn't affect exposure. The amount of light falling on the subject is the determining factor.

 

[dennybeall]

I agree with the conclusions Ysarex made with his experiment but would like to see the same experiment with 5 shots - 2 up close door panels, and the original shots with 2 door panel spot metered and the original metering.
How would the 2 sets of door panels compare?

 

[andytakeone]

I made the above illustration one afternoon for my class -- using a reflected light meter built into a camera. There are three photos there; The top scene shows two sedans parked across the street from my house. Below the silver sedan is a section of the door panel directly below the door handle. Below the charcoal sedan is a section of the door panel directly below the door handle.

First I took the photo of the entire scene including both cars using the camera meter's exposure determination. One car reflects more light than the other. With both cars in the scene the meter averages the reflectance of both. Then I walked across the street and got real close and photographed the silver car door panel -- then the charcoal car door panel. In these two photos I again used the camera meter's exposure determination. ALL THREE EXPOSURES WERE DIFFERENT. The exposure for the silver car door panel was less than the overall scene and the exposure for the charcoal door panel was more than the overall scene. The camera meter darkened the silver car's door panel and lightened the charcoal car's door panel.

Open the photo in an app like PS or GIMP where you can measure HSB (hue, saturation, brightness) values and check those four spots. In the top photo I get a brightness value for the silver car door: 53 and for the charcoal car door: 16 -- the point: that's a substantial difference as it should be -- the silver car reflects more light and is brighter. In the two closeups of the door panels I get silver car door: 38 and charcoal car door: 39.

Joe

Hmm. So if I'm not mistaken, this is similar to the instance of shooting in bright snow where the camera's meter attempts to move the white snow to middle grey and so you have to compensate the exposure.

Thanks everyone for their replies, but I have to admit I'm still a little muddled.

The main point of contention is... for example:
I have this EV chart that says EV15 is for a clear sunny day (Sunny 16), but that EV16 is also a clear sunny day, except with bright snow or sand.
This leads me to believe that the reflected light coming off the snow or sand affects the exposure in that one must adjust to the added light (from the "reflectance") by stopping down.

Then, if I extrapolate this information, almost any scene where there are objects that reflect a lot of light (white surfaces for example), I must not only take into consideration the current weather and light, but also the reflecting light coming off the objects (incident light and reflectance vs. just incident light).

So, going back to the snow-middle grey compensation, or the silver and charcoal cars. Is the meter's darkening of the silver car and brightening of the charcoal car due to a factor of color and tone? As in it's trying to get it to middle grey? (not to mention the snow). That makes it seem like it doesn't have to do with light, but with... colors.

 

[Dave442]

Just a thought, what if you check the difference in exposure your camera gives off of a reflective surface like green plant leaves when using a circular polariser and comparing the readings with the polariser set to let the "glare" through and then set to remove the reflectance.

 

[Ornello]

I made the above illustration one afternoon for my class -- using a reflected light meter built into a camera. There are three photos there; The top scene shows two sedans parked across the street from my house. Below the silver sedan is a section of the door panel directly below the door handle. Below the charcoal sedan is a section of the door panel directly below the door handle.

First I took the photo of the entire scene including both cars using the camera meter's exposure determination. One car reflects more light than the other. With both cars in the scene the meter averages the reflectance of both. Then I walked across the street and got real close and photographed the silver car door panel -- then the charcoal car door panel. In these two photos I again used the camera meter's exposure determination. ALL THREE EXPOSURES WERE DIFFERENT. The exposure for the silver car door panel was less than the overall scene and the exposure for the charcoal door panel was more than the overall scene. The camera meter darkened the silver car's door panel and lightened the charcoal car's door panel.

Open the photo in an app like PS or GIMP where you can measure HSB (hue, saturation, brightness) values and check those four spots. In the top photo I get a brightness value for the silver car door: 53 and for the charcoal car door: 16 -- the point: that's a substantial difference as it should be -- the silver car reflects more light and is brighter. In the two closeups of the door panels I get silver car door: 38 and charcoal car door: 39.

Joe

Hmm. So if I'm not mistaken, this is similar to the instance of shooting in bright snow where the camera's meter attempts to move the white snow to middle grey and so you have to compensate the exposure.

Thanks everyone for their replies, but I have to admit I'm still a little muddled.

The main point of contention is... for example:
I have this EV chart that says EV15 is for a clear sunny day (Sunny 16), but that EV16 is also a clear sunny day, except with bright snow or sand.
This leads me to believe that the reflected light coming off the snow or sand affects the exposure in that one must adjust to the added light (from the "reflectance") by stopping down.

Then, if I extrapolate this information, almost any scene where there are objects that reflect a lot of light (white surfaces for example), I must not only take into consideration the current weather and light, but also the reflecting light coming off the objects (incident light and reflectance vs. just incident light).

So, going back to the snow-middle grey compensation, or the silver and charcoal cars. Is the meter's darkening of the silver car and brightening of the charcoal car due to a factor of color and tone? As in it's trying to get it to middle grey? (not to mention the snow). That makes it seem like it doesn't have to do with light, but with... colors.

In the case of sand or snow, the issue is that light is reflected around the scene more than usual, so the intensity of light falling on the scene is slightly greater.

 

[Ysarex]

Hmm. So if I'm not mistaken, this is similar to the instance of shooting in bright snow where the camera's meter attempts to move the white snow to middle grey and so you have to compensate the exposure.

Correct. We normally point the camera at objects rather than lights. Objects reflect light and so the camera meter is reading reflected light coming off the objects. Different objects reflect light at different percentage rates. Look at this photo of a wild geranium:

The white flower petals and green leaves are in the sun. Fallen leaves on the ground are in sun and shade. The white flower petals reflect more light (higher percentage) than either the green leaves or the fallen leaves. The fallen leaves in the shade reflect back even less light. The camera meter is a measuring device -- that's all. It doesn't know if this is a flower or a birthday cake or a train engine. It doesn't know if it's the flower you want to photograph or the fallen leaves on the ground that you're after. It just does it's job and measures the intensity of the (reflected) light it's exposed to.

You have some setting options you can use with the tool. Restrict the field of vision (spot) or disproportionately weight the area being measured yet still accounting for the entire frame area. It's up to you to use the tool and only you have evaluative capability (despite what Canon suggests with their metering mode).

THE ASSUMPTION: This is a fair assumption -- when you take all of the objects in the frame that are reflecting light at different rates (white petals are brighter than green leaves), and you average those reflectance rates together you usually get an average close to 18% reflectance. When the measurement data is returned from the meter and used to determine exposure settings that assumption applies. If your evaluation and experience determine otherwise you must intervene if you don't want something that was and should be light to be rendered darker or something that was and should be dark to be rendered lighter.

Thanks everyone for their replies, but I have to admit I'm still a little muddled.

The main point of contention is... for example:
I have this EV chart that says EV15 is for a clear sunny day (Sunny 16), but that EV16 is also a clear sunny day, except with bright snow or sand.

Get a new EV chart. For ISO 100 speed film or digital sensor with a subject in direct daytime sunshine the EV for correct exposure is 15. That's EV 15 for a coal pile, a forest, city buildings, the beach and the ski slopes.

This leads me to believe that the reflected light coming off the snow or sand affects the exposure in that one must adjust to the added light (from the "reflectance") by stopping down.

No.

Then, if I extrapolate this information, almost any scene where there are objects that reflect a lot of light (white surfaces for example), I must not only take into consideration the current weather and light, but also the reflecting light coming off the objects (incident light and reflectance vs. just incident light).

So, going back to the snow-middle grey compensation, or the silver and charcoal cars. Is the meter's darkening of the silver car and brightening of the charcoal car due to a factor of color and tone? As in it's trying to get it to middle grey? (not to mention the snow). That makes it seem like it doesn't have to do with light, but with... colors.

There is no color without light. The color that you perceive an object to be is the result of the light reflecting from it. Your eye perceives color in response to light stimulation. That car example would have worked exactly the same if the silver car were yellow and the charcoal car were blue -- or pink and purple or lime and brown.

Joe

 

[snowbear]

This leads me to believe that the reflected light coming off the snow or sand affects the exposure in that one must adjust to the added light (from the "reflectance") by stopping down.

Actually, the opposite is true, but not because of reflection. In snow (a lot of snow), you typically overexpose by a stop or stop-and-a-half, otherwise you get grey snow.

 

[andytakeone]

Hmm. So if I'm not mistaken, this is similar to the instance of shooting in bright snow where the camera's meter attempts to move the white snow to middle grey and so you have to compensate the exposure.

Correct. We normally point the camera at objects rather than lights. Objects reflect light and so the camera meter is reading reflected light coming off the objects. Different objects reflect light at different percentage rates. Look at this photo of a wild geranium:

The white flower petals and green leaves are in the sun. Fallen leaves on the ground are in sun and shade. The white flower petals reflect more light (higher percentage) than either the green leaves or the fallen leaves. The fallen leaves in the shade reflect back even less light. The camera meter is a measuring device -- that's all. It doesn't know if this is a flower or a birthday cake or a train engine. It doesn't know if it's the flower you want to photograph or the fallen leaves on the ground that you're after. It just does it's job and measures the intensity of the (reflected) light it's exposed to.

You have some setting options you can use with the tool. Restrict the field of vision (spot) or disproportionately weight the area being measured yet still accounting for the entire frame area. It's up to you to use the tool and only you have evaluative capability (despite what Canon suggests with their metering mode).

THE ASSUMPTION: This is a fair assumption -- when you take all of the objects in the frame that are reflecting light at different rates (white petals are brighter than green leaves), and you average those reflectance rates together you usually get an average close to 18% reflectance. When the measurement data is returned from the meter and used to determine exposure settings that assumption applies. If your evaluation and experience determine otherwise you must intervene if you don't want something that was and should be light to be rendered darker or something that was and should be dark to be rendered lighter.

Thanks everyone for their replies, but I have to admit I'm still a little muddled.

The main point of contention is... for example:
I have this EV chart that says EV15 is for a clear sunny day (Sunny 16), but that EV16 is also a clear sunny day, except with bright snow or sand.

Get a new EV chart. For ISO 100 speed film or digital sensor with a subject in direct daytime sunshine the EV for correct exposure is 15. That's EV 15 for a coal pile, a forest, city buildings, the beach and the ski slopes.

This leads me to believe that the reflected light coming off the snow or sand affects the exposure in that one must adjust to the added light (from the "reflectance") by stopping down.

No.

Then, if I extrapolate this information, almost any scene where there are objects that reflect a lot of light (white surfaces for example), I must not only take into consideration the current weather and light, but also the reflecting light coming off the objects (incident light and reflectance vs. just incident light).

So, going back to the snow-middle grey compensation, or the silver and charcoal cars. Is the meter's darkening of the silver car and brightening of the charcoal car due to a factor of color and tone? As in it's trying to get it to middle grey? (not to mention the snow). That makes it seem like it doesn't have to do with light, but with... colors.

There is no color without light. The color that you perceive an object to be is the result of the light reflecting from it. Your eye perceives color in response to light stimulation. That car example would have worked exactly the same if the silver car were yellow and the charcoal car were blue -- or pink and purple or lime and brown.

Joe

Hmm, I see. Thanks for this comprehensive reply.

One last thing though and then I think I got it:

You say that in the photo with the white geranium the assumption applies that the light meter is taking into account the light reflecting off the geranium's surface--that a mixture of the less reflective objects in the photo (i.e. the green leaves in the shade) with the more reflective geranium, averages out the overall reflectance to about 18%. (I suppose given that the camera is in an evaluative metering mode).

Now, say we fill the entire frame with white geraniums, to the point that we no longer see shade. Averaged out, let's say that boosts the photo's overall reflectance to 50%. Does that mean we must stop down to compensate for the added light? The incident light hasn't changed, but more light has been introduced with more reflective white geraniums. And if this is true, doesn't that mean when one shoots a snowy landscape on a clear day, that the overall reflectance is also higher (making the difference between EV15 and EV16)?