Digital imaging processing for in-vivo skin measurements

In this paper, we have proposed a new method for extracting melanin and hemoglobin information from a single human skin digital colour image, based on the skin spectra information and digital camera sensor spectra sensitivity profiles. The digital skin colour image was first separated into RGB images, then based on different contributions of melanin and hemoglobin at different red, green and blue spectra regions, we have managed to reconstruct melanin image and hemoglobin image from the original RGB images. We will first present the theoretical background, then the analysis results of digital colour images of palm, forearm, and face.

[1]  R. Anderson,et al.  The optics of human skin. , 1981, The Journal of investigative dermatology.

[2]  K. Grammer,et al.  Color homogeneity and visual perception of age, health, and attractiveness of female facial skin. , 2007, Journal of the American Academy of Dermatology.

[3]  Norimichi Tsumura,et al.  Mapping Pigmentation in Human Skin by Multi-Visible-Spectral Imaging by Inverse Optical Scattering Technique , 2000, Color Imaging Conference.

[4]  S. Nayar,et al.  The Appearance of Human Skin , 2005 .

[5]  Norimichi Tsumura,et al.  Image-based skin color and texture analysis/synthesis by extracting hemoglobin and melanin information in the skin , 2003, ACM Trans. Graph..

[6]  H.J.C.M. Sterenborg,et al.  Skin optics , 1989, IEEE Transactions on Biomedical Engineering.