Title :
Face detection based on skin mask, number of holes and SVM
Author :
Ghaderi, N. ; Ghaderi, A.H.
Author_Institution :
Fac. of Eng., Shahrekord Univ., Sharekord, Iran
Abstract :
This paper introduces a new 3-step face detection algorithm using MATLAB software. This 3-step algorithm is based on skin mask, number of holes in each region and Support Vector Machine (SVM) for face detection in RGB images. Due to the use of just 10 levels images in the proposed algorithm, the SVM training step is accomplished so fast. In addition, it will be shown that this method is so accurate in comparison with the previous methods, because of using these 10 levels just around the optimum threshold value.
Keywords :
face recognition; image colour analysis; learning (artificial intelligence); mathematics computing; skin; support vector machines; 3-step face detection algorithm; MATLAB software; RGB images; SVM training step; number of holes; optimum threshold value; skin mask; support vector machine; Face Detection; Number of Holes; SVM; Skin Color;
Conference_Titel :
Information Science and Control Engineering 2012 (ICISCE 2012), IET International Conference on
Conference_Location :
Shenzhen
Electronic_ISBN :
978-1-84919-641-3
DOI :
10.1049/cp.2012.2374
