ABSTRACT
In this paper, we provide a thorough review of color based skin segmentation methods. A selection of the best documented methods found in the state of the art has been made in order to get a consensus-like decision making system which has been used as the reference skin classification system to train a multilayer perceptron (MLP) artificial neural network (ANN) in order to map color information into skin/non-skin class information. Results prove that this integration clearly provides a lower error rate for color based skin segmentation than any of the individual methods, thus offering a promising alternative for fast, accurate and robust skin segmentation ideal both for real time image processing environments and adaptive skin tracking solutions.
