Skin Hemoglobin and Melanin Quantification on Multi-Spectral Images

Hao Gong and Michel Desvignes

Keywords

multi-spectral images, hemoglobin, melanin, NMF, model-fitting, graph cuts

Abstract

In this paper, we propose and compare two different approaches for quantification of skin hemoglobin and melanin on multi-spectral images. The first method is based on non-negative matrix factorization (NMF) with multiplicative update algorithm. The second method is a Beer-Lambert law based model-fitting technique. Quantitative evaluation through graph-cut segmentation on melanoma indicates that model-fitting method obtains more accurate quantification than NMF.

Important Links:

• DOI: 10.2316/P.2012.771-039
• From Proceeding (771) Imaging and Signal Processing in Health Care and Technology / 772: Human-Computer Interaction / 773: Communication, Internet and Information Technology - 2012

Go Back