; ProgressOn-07-17-2012
Documents
User Generated
Resources
Learning Center
Your Federal Quarterly Tax Payments are due April 15th

# ProgressOn-07-17-2012

VIEWS: 0 PAGES: 1

• pg 1
```									s_skinReflectance2AbsorbanceParameters.m (in pdcprojects/Skin/Code/Joyce)

The purpose of this script is to validate the linear model we use for skin absorbance. We do this by
showing that the linear model can predict measured skin absorbances. We convert skin reflectance to
skin absorbance and then use a linear that predicts the absorbance data following Yudovsky et al 2010

The idea is that the measured reflectance is
Ref(l) = 10.^ -(w1 Hb0(l) + w2 Hb(l) + w3 M(l) + K)
So,
-log10(Ref(l)) = (w1 Hb0(l) + w2 Hb(l) + w3 M(l) + K)

We estimate the relative concentration of oxy, deoxy and melanin and an additive factor for scatter

Requires SkinComponentAbsorbances.mat in … iset-4.0\data\surfaces\absorbances
And from iset/data/surfaces/reflectances:
 CaucasianHyspex
 AsianHyspex
 EastIndianSkinReflectances
 MiddleEastSkinReflectances
 AfricanAmericanSkinReflectances

s_skinSamples.m (in pdcprojects/Skin/Code/Joyce)

Having validated the linear model for skin absorbances, we create simulated skin reflectances by varying
the relative weights on oxy, deoxy, melanin and scatter. The skin reflectances are defined by three
parameters
1. Oxy (deoxy = 1-oxy)
2. Melanin
3. Scatter

Requires SkinComponentAbsorbances.mat in … iset-4.0\data\surfaces\absorbances

s_sensorPredictSkinComponentsJF.m (in svn/pdcprojects/skin/code/Scratch)

We create a reflectance chart with the simulated skin reflectances generated by s_skinSamples.m (each
associated with 3 stimulus parameters: oxy, melanin and scatter)
We predict the sensor values for these scene data
We then find a mapping of sensor values into the stimulus space

requires SkinSimulations.mat and SimSkinReflectanceData.mat in
… pdcprojects\Skin\Code\Scratch

```
To top