Human skin color code recognition: A case study

Abstract

Human skin color has been studied as biometric indicator, most of previous researches studies focused in using skin color to detect the face, human and human motion. The objective of this article is to evaluate the variation in human skin color as a unique code for identification purpose. Fifty picture of the nose were captured using a mobile camera with a resolution of 3 mega pixel. Part of the nose area has been the region of interest to represent a region of less direct exposure to the sun light as the rest of the face. The color images were converted from RGB to HSI format in order to isolate the effect of the light intensity during shots. The images were analyzed using mathematical and statistical methods (Mean, Median, standard deviation, kurtosis, skewness, and gray level co-occurrence matrices (GLCM). This algorithm found that the variation in skin color from one person to another is statistically remarkable when implementing this technique. © 2013 IEEE.