A new and improves skin detection method using RGB vector space

Abstract

This paper describes a new method for skin detection based on RGB vector space. Skin color has proven to be a useful cue for pre-process of face detection, localization and tracking. Image content filtering, content aware video compression and image color balancing applications can also benefit from automatic detection of skin in images. Numerous works for skin color segmentation and detection have been reported in the literature in the last few years. In this paper first few pixel based methods are presented and then vector calculations as an efficient approach for determination of skin area based on implicit mathematical model is proposed. In order to evaluate our method we performed an experimental study. Results obtained are very promising and shows important advantages over other techniques.