Threshold Based Skin Color Classification

Authors

  • Sasan Karamizadeh Advanced Informatics School (AIS), Universiti Teknologi Malaysia (UTM), Kuala Lumpur, 54100, Malaysia.
  • Shahidan M Abdullah Advanced Informatics School (AIS), Universiti Teknologi Malaysia (UTM), Kuala Lumpur, 54100, Malaysia.
  • Jafar Shayan Advanced Informatics School (AIS), Universiti Teknologi Malaysia (UTM), Kuala Lumpur, 54100, Malaysia.
  • Parham Nooralishahi Department of Computer Science and Information Technology, University of Malaya (UM), Kuala Lumpur, 50603, Malaysia
  • Behnaz Bagherian Faculty of Computer Science and Information Technology, University Putra Malaysia (UPM).

Keywords:

Skin Segmentation, Image Noise, K-NN, Multi Skin,

Abstract

In this paper, we presented a new formula for skin classification. The proposed formula can overcome sensitivity to noise. Our approach was based multi-skin color Hue, Saturation, and Value color space and multi-level segmentation. Skin regions were extracted using three skin color classes, namely the Caucasoid, Mongolid and Nigroud. Moreover, in this formula, we adopted Gaussian-based weight k-NN algorithm for skin classification. The experiment result shows that the best result was achieved for Caucasoid class with 84.29 percent fmeasure.

Downloads

Published

2017-06-01

How to Cite

Karamizadeh, S., M Abdullah, S., Shayan, J., Nooralishahi, P., & Bagherian, B. (2017). Threshold Based Skin Color Classification. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 9(2-3), 131–134. Retrieved from https://jtec.utem.edu.my/jtec/article/view/2341