Consider a small region of interest captured by all imaging modalities. Use the color sampler tool with a sample size of 31 by 31 average. Enter L* values for each wavelength band to generate the bar chart.

*This method uses perceptual L* data (CIELAB). It calculates reflectance from percieved brightness levels (human vision-based conversion). This method gives mathematically derived reflectance based on simplified perceptual data. It does not involve sensor calibration, which would provide linear sensor behavior. 

-------
Multiband imaging isn’t typically calibrated like multispectral and hyperspectral imaging. However, you could in theory calibrate the results to provide a more accurate calculation. 

Raw image: pixel values from image of the object. Neutralize all adjustments in camera raw, set color space to ProPhoto RGB or Adobe RGB, 16 bits/channel.

Dark reference: pixel values from image captured with the lens cap on, same exposure time as Raw to record any sensor dark curent and baseline noise.

White reference: pixel values from image of lambertian reflectance standard (99% Spectralon), to record sensor response to known reference. 

---
Step by step sensor-based calibration method:

This example is for cobalt blue:

• Raw Image RGB: (79, 109, 169)
• Raw Image IR:   (233, 233, 233)
• Raw Image UVR: (187, 187, 187)
• Dark Reference RGB: (0, 0, 0) - a healthy sensor will read 0s.
• White Reference RGB: (244, 244, 244) - properly calibrated should be same # across channels.
  

You’d calculate reflectance separately for each RGB channel:

Red channel reflectance (%):

Green channel reflectance (%):    

Blue channel reflectance (%):



Differences in calculation results, same colorant reference - 

Response Vector Chart approximate L* values:
Measures how humans perceive brightness