Improved Dynamic Threshold Method for Skin Colour Detection Using Multi-Colour Space
Mohd Zamri Osman, Mohd Aizaini Maarof and Mohd Foad Rohani
DOI : 10.3844/ajassp.2016.135.144
American Journal of Applied Sciences
Volume 13, Issue 2
This paper presents a skin colour detection based on animproved dynamic threshold method to reduce false skin detection. Current fixedthreshold skin detection fails in certain situations such as misclassificationbetween non skin-like with similar skin-like colour. Any true skin may falsely bedetected as non-skin. Research work introduces high-level skin detection strategybased on online sampling where offline training is not required. This strategyshows a promising performance in term of classifying images under skin-like andethnicity image variations. However, some of the methods produced high falsepositives that reduced the accuracy of skin detection performance. Therefore,in this study, instead of single colour space and fixed threshold method, animproved skin detection based on multi-colour spaces is proposed. Furthermore,a dynamic threshold method also has been improved by introducing elasticelliptical mask model for online skin sampling. The experimental result showsan improvement in employing multi-colour rather than single colour space byreducing the false positive and increasing the precision rate.
© 2016 Mohd Zamri Osman, Mohd Aizaini Maarof and Mohd Foad Rohani. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.