Skin Detection Based on Image Color Segmentation with Histogram and K-Means Clustering

Skin detection is a crucial pre-processing step for finding human faces in images. The challenging task is to find a reliable, yet efficient method to detect skin region(s). In this paper, we propose a new, simple and efficient method for skin detection based on image segmentation of different color spaces, and simple clustering technique (K-means) to cluster similar pixels on an image. As input features, K-means clustering uses two components from two different color spaces (Hue, Cr, Cb), positions of pixels on an image and rough estimation of skin pixels obtained from skin-color based detection. Our approach shows promising results on human images from different ethnicities, with simple background and high illumination. It has low computational cost, since it does not require any training. Results indicate that the method is suitable as a pre-processing step for some supervised method for advanced human skin segmentation and detection.