How can you have a scene linear reflectance greater than 100%?

Started 2 months ago | Questions thread
OP Mandem Junior Member • Posts: 28
Re: Diffuse = max 100%, Mixed and Specular > 100%, Linearly
1

Jack Hogan wrote:

Mandem wrote:

alanr0 wrote:

MrBrightSide wrote:

Those numbers from 0 to 720 look like mV values for the luminance channel. Is that what you’re asking about?

Anything is possible, but the OP is convinced they represent percentage reflectance values - possibly because that is how they are labelled on the screen shot.

The choice of 18% and 90% reflectance values seems significant. These numbers crop up in the ITU documents that spider-mario linked, including ITU report BT.2408-4, where they appear as widely used grey scale reflectance values. See for instance the figure here.

You could check out the gitbub repository for LUTCALC which I believe is the source of the graphic. My guess is careless terminology for relative exposure values, but I have not dug deeper to find out.

Quite the contrary. After our discussion, I am fully convinced the "reflectance" values on Lutcalc are indeed luminance values. The idea of an 18% or 90% is in reference to the luminance of an 18% reflectance gray card or 90% reflectance card. Like you yourself said, you can't have a reflectance greater than 100%.

If you are a practitioner and you've followed posts in this thread, you should by now understand that reflectance comes in three flavors: diffuse, specular and mixed diffuse and specular.

Maximum diffuse white reflection in photography is considered to be 100%. From the ITU link above:

"Reference Level: HDR Reference White (100%) also diffuse white and Graphics White"

However, there is often detail above 100% diffuse white which one may want to capture, as explained in the earlier post on Ansel Adams, clouds and snow. Such detail is typically the result of mixed reflections which live above 100%. In some extreme cases one may actually want to capture the illuminant source itself (say the sun), in which case we would be capturing a signal several thousand % above maximum diffuse white. All this linearly, thanks to the physics of radiometry.

Photography normally works in relative units, that's why many people don't know at what absolute exposure their cameras clip in lx-s - but are interested in where an 18% gray card fits on the histogram in order to choose the appropriate relative aperture and exposure time to capture all the detail they are interested in.

Absolute values of reflected Radiance or Luminance in photography are mainly relevant when perceptual phenomena need to be taken into consideration, say for instance in for adaptation and HDR applications.

Jack

I'm not a practitioner. Just someone who is genuinely interested. But yes, I have grasped the concept of Diffuse, Specular, and Mixed. If you look at my example above with the camera out in a field at noon directly facing a 100% diffuse reflector and a mirror, you'll see that's exactly where I take into consideration the nature of diffuse and specular reflectance. Both of the objects are reflecting a 100% of the light(Theoretically a perfect mirror and a perfect diffuse reflector). You just can't get more than a 100% reflectance according to the formulas Alan put up(exception being Brighteners but they're a rarity in the natural world):

Reflected/Incident Luminous Flux

Reflected/Incident Luminous Energy.

The fundamental difference lies within the DIRECTIONALITY of the light rays being reflected (Hence what makes something Diffuse or Specular or Mixed). Of course, a mirror reflecting will have much stronger Luminance Values than the 100% diffuse reflector purely due to there being no scattering of the rays, even though they both reflect a 100% of the light

Again this all comes back to trying to understand why there were values greater than a 100% on an x-axis called "reflectance" In that LutCALC Graph. It just didn't make any sense since, again, you can't get a value bigger than 1 on the formulas Alan posted above. Which is why the idea that "reflectance" actually refers to LUMINANCE is more fitting. When it says 18% reflectance or 90% reflectance or 100% reflectance in that LutCALC graph I believe it's referring to the Luminance of an 18%,90% or 100% diffuse reflector. The fact that it is LUMINANCE is also the reason why, I believe, the values can go above a 100% because certain reflectors(specular or mixed) or even incident light sources due to their directionality can be significantly more intense than a 100% diffuse reflector. Hope this made sense.

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow