
Skin Reflectance Model for Multispectral System
Author(s) -
Feng Sui,
Jinjin Zheng
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1550/2/022037
Subject(s) - multispectral image , reflectivity , wavelength , materials science , perpendicular , multispectral pattern recognition , optics , remote sensing , geology , optoelectronics , mathematics , physics , geometry
This is a pre-study on the non-invasive detection and contrast observation of pigmented lesions. We obtained a proper skin reflectance model of a multispectral imaging system that generates images perpendicular exposure of skin perpendicularly exposed reflectance in the wavelength ranging from 470 to 630 nm at an internal of 2 nm, altogether 81 different wavelengths in the visible range. The reflectance data was collected form 27 different skin lesions each 500*500 pixels in various conditions from 8 persons. Since the spectrum is mixture of layers of skin tissue distribution, independent component analysis (ICA) and pure components analysis were proposed to fit the experimental data, and the layered-dependent skin parameters melanin content, blood volume fraction, and haemoglobin content were extracted from the skin reflectance model. The skin reflectance model fitted spectrum curves show 4% average (12% maximum) difference from the experimental data, which were acceptable. The skin model is accepted for the multispectral imaging system in further studies.