luminance matching

JA
Posted By
Jasper_Arends
Feb 3, 2009
Views
633
Replies
6
Status
Closed
Hi,

I have 18 pictures that I want to have, for research purposes, EXACTLY the same luminance. The pictures are cut out faces and the background needs to stay white. How can I do this?

thanks a lot!

Mariska
Related Tags
luminance match

How to Master Sharpening in Photoshop

Give your photos a professional finish with sharpening in Photoshop. Learn to enhance details, create contrast, and prepare your images for print, web, and social media.

MK
Mariska_Kret
Feb 3, 2009
Hi,

If this is not possible, I have another question: can I somehow calculate the mean/average luminance of a picture? In that case I can compare my pictures and see which ones are really deviant from the rest.

Mariska
GH
Gernot_Hoffmann
Feb 3, 2009
Mariska,

here are some hints:

1. Color space: Lab .
2. Measuring: Histogram, channel L, mean value.
Don’t forget to update the calculation, mainly
if a warning appears.
3. Area: either the whole image or a selection
of the face.
4. Editing: Because of the available selection it’s
not difficult to keep the background white and to
adjust the lightness of the face by a Curve (L).

More advanced methods which don’t require accurate
selections are explained here:

<http://www.fho-emden.de/~hoffmann/labproof15092008.pdf>

Best regards –Gernot Hoffmann
JM
J_Maloney
Feb 3, 2009
Blur… average and you can set the color sampler to show LAB and then watch the L. Or set the sampler to 101 x 101 and just scrub around.
GA
George_Austin
Feb 4, 2009
Mariska

What with the loose usage of "luminance" in digital photography, one has to regard the term as ambiguous until further defined. Do you seek to keep constant the luminosity (a weighted sum of R,G,B values), the luminance as in HSL color space, the Lightness as in LAB color space, the brightness as in HSB space, the physical luminance (intensity), etc etc.

When you have decided which version of luminance you need to nurse, you need to specify whether you want point-to-point preservation or overall average and, if the latter, including or excluding the white background.
MK
Mariska_Kret
Feb 4, 2009
Hi!

Thank you for the replies! I’m very happy that there is a solution!

As I’m going to use eyetracking, the most important thing for me is that the pupil size won’t change because of luminance (darkness) differences between different stimuli.
George, I don’t know the difference between all the things you mention. I’m afraid it’s not really my field. For sure the background may not be changed as this can invoke saccades to the background and this is not what I want. So I think the solution of Gernot and Maloney is the best thing, right? But can you please explain me a bit more how to do it? I’m actually completely new to photoshop.

greetings,
Mariska
GA
George_Austin
Feb 4, 2009
Mariska,

Assuming the faces occupy about the same fractional area of the image being viewed and are in about the same position, the pupil dilation should be the same as long as any one of the various "luminances" is constant. The histogram data includes an average luminosity using weighted rgb values, and that should suffice for your purposes, although it’s not clear what precision you need. Adjust each image to give the same average luminosity as read from the histogram’s data.

How to Improve Photoshop Performance

Learn how to optimize Photoshop for maximum speed, troubleshoot common issues, and keep your projects organized so that you can work faster than ever before!

Related Discussion Topics

Nice and short text about related topics in discussion sections