Simulated reflectance spectra and point spread functions in database constructed by moderate grouping of nine layers in skin model

Abstract

This paper presents a numerical study on simulated spectra and point spread functions in the database that we constructed by moderate grouping of nine layers in a skin model. The database is used for identification of a simulated skin reflectance spectrum which seems most closed to a certain target spectrum or a measured spectrum. On the basis of the properties of absorption and scattering coefficients in a nine-layered skin tissue model, we classify nine layers into two and three groups for changing values of the absorption and scattering coefficients, respectively, and then reflectance spectra are simulated by Monte Carlo method with these five groups to build reflectance spectral database. The desired spectrum is extracted from the database by using the root mean square error. To identify uniquely the most desirable spectrum among some similar spectra contained in the database, we propose additionally the use of a point spread function of reflected intensity on the skin surface. Simulated results demonstrate that this approach is promising to identify the most likely spectrum that is fit to the target spectrum. The finally identified spectrum in the database gives its absorption and scattering coefficients in each of the nine layers which are used to estimate skin conditions.