Title :
Face detection based on chrominance and luminance for simple design
Author :
YoungJin Kim ; Sungkwang Cho ; Byungjoon Back ; TaeChan Kim
Author_Institution :
Image Dev. Team, Samsung Electron., Yongin, South Korea
Abstract :
We proposed improved face detection system using Grid of Skin Color Regions besides luminance data that are used in the conventional system. It decreases FAR (False Acceptance Ratio) in image part where there are non-skin color. Using skin information in CbCr plane is proposed to be implemented into embedded system. It separates skin color and non-skin color regions in the several regions. After that, AdaBoost algorithm is used in training face and non-face images. We confirmed the result that brings less FAR compared to the conventional method.
Keywords :
brightness; embedded systems; face recognition; image colour analysis; learning (artificial intelligence); object detection; AdaBoost algorithm; CbCr plane; FAR; chrominance; embedded system; face detection; false acceptance ratio; luminance; nonskin color regions; Databases; Face; Face detection; Image color analysis; Skin; Training; Training data; AdaBoost; ICT; chrominance; face detection; face recognition;
Conference_Titel :
SoC Design Conference (ISOCC), 2012 International
Conference_Location :
Jeju Island
Print_ISBN :
978-1-4673-2989-7
Electronic_ISBN :
978-1-4673-2988-0
DOI :
10.1109/ISOCC.2012.6407103
