A novel skin detection method using Generalized Discriminant Analysis

Abstract

In this work, a novel method for skin detection in color images is proposed. For this purpose, three different types of kernel functions are applied within the framework of the Generalized Discriminant Analysis (GDA) algorithm in the YCbCr color space. The performance of the resulting system is evaluated using the Compaq database. Our experiments show that by adopting the proposed approach, the correct detection rate of the skin detection system will improve.