Hi !
I am undergraduate student helping a doctorate student on her project. We want to measure the color of jackpine seeds across its range. We took the pictures using a full-spectrum converted Canon EORS RP 2019, Kolari Vision Flash as a light source and used ColorChecker passport Photo 2 as standards. Pictures did not look overexposed or underexposed at the time we took them. However, when we convert to multispectral image and measure the seeds’ reflectance values (by pressing m for the ROI), some values are over 100%. What could potentially explain those >100% values (light source, non-uniform lighting in the pictures, wrong values of the standards) ? If we want to compare the seed colors between individuals or populations across the range. does it represent a problem ? If it is a problem, how can we solve it ?
Thank you very much
Hi,
Pixels with more than 100% reflectance possibly indicate a mismatch between your standard values provided to the toolbox and the ‘physical reality’ of your image. It could be that the standard values you provide are not accurate or that the lighting between the standard and parts of your image is not homogeneous.
Cheers,
Cedric