Skin Segmentation by Using Complex Valued Neural Network with HSV Color Spaces

Authors : Eda Capa Kiziltas, Ayşenur Uzun, Ersen Yilmaz

Pages : 1-4

View : 20 | Download : 6

Publication Date : 2019-03-04

Article Type : Research

Abstract :Nowadays, digital image processing is useful tool in daily life for security, surveillance and artificial intelligence applications. Mainly in nudity alerts and face detection, skin segmentation is widely preferred method due to its simple background. The main problem in skin segmentation is skin color variation that result of different percentage of pigments, number and size of melanin particles in human-being. In literature, there are rule-based and hybrid models for skin segmentation, however rule-based algorithm are not enough to overcome skin variety. As hybrid mode Complex valued neural network (CVNN) and color space transformation is applied to Skin Segmentation Database from UCI Learning Repository that is collection of different age and race group human’s skin samples in RGB format. 
Keywords : Skin Segmentation, Complex valued neural network (CVNN), Machine Learning