Abstract
AbstractSkin color detection is a challenging problem in the field of image processing. Basically, the color of human skin is created by a combination of blood (red) and melanin (yellow, brown). Skin colors lie between these two extreme hues and are somewhat saturated. The main purpose of this proposed algorithm is to improve the detection and recognition of human skin color. Skin color detection process is useful in a wide range of image processing applications such as diagnosis in dermatology or skin disease, face recognition for security purpose, gesture tracking, and computer-human interaction. Different color systems existed for various applications. Color model is a mathematical representation of colors which holds the color space with its primary color’s components like red, green, and blue. We used different color models to obtain suitable result for detection of human skin color. Furthermore, the proposed approach is based on a threshold value and skin detection has been performed on different color models. The pixel-based technique is used in skin detection which segregates skin and non-skin pixels. The experimental results are obtained using human skin color detection technique with different color space such as RGB, HSV, YCbCr, and CMYK.KeywordsSkin detectionColor image processingColor modelsRGBHSVYCbCrCMYKFull colorPseudo-color
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call