Paper
1 October 2011 A new color space to discriminate skin regions using fuzzy quadratic programming
Amir Mazaherie, Saeed Mirghasemi, Mojtaba Lotfizad
Author Affiliations +
Proceedings Volume 8285, International Conference on Graphic and Image Processing (ICGIP 2011); 82853R (2011) https://doi.org/10.1117/12.914447
Event: 2011 International Conference on Graphic and Image Processing, 2011, Cairo, Egypt
Abstract
This paper presents a new color space through linear conversion in which the cost function has fuzzy quadratic (FQ) characteristic. Training samples are used to create an efficient color space with respect to desired application. Quadratic Programming (QP) technique is engaged to extract the conversion matrix. According to fuzzy feature of the cost function, several matrices are introduced to generate the new color space for skin segmentation. The proposed method involves two steps. First a second order cost function established based on separating target pixels from non-target ones while followed by some constraints. In the second stage, a quadratic optimization is used to solve problem efficiently. Finally, the results of fuzzy approach are compared to non-fuzzy methods.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Amir Mazaherie, Saeed Mirghasemi, and Mojtaba Lotfizad "A new color space to discriminate skin regions using fuzzy quadratic programming", Proc. SPIE 8285, International Conference on Graphic and Image Processing (ICGIP 2011), 82853R (1 October 2011); https://doi.org/10.1117/12.914447
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Skin

Matrices

RGB color model

Computer programming

Fuzzy logic

Facial recognition systems

Evolutionary algorithms

RELATED CONTENT


Back to Top