Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Journal of Jiangxi Normal University(Natural Science Edition)》 2013-04
Add to Favorite Get Latest Update

Face Detection Algorithm Based on Double Color Space and Facial Rectangular Characteristics

ZHU Zhi-liang;XIONG Feng;TAO Xiang-yang;LIU Xiao-shan;College of Physics & Communications Electronics,Jiangxi Normal University;Key Laboratory of Photo-electronic & Telecommunication of Jiangxi Province;  
An algorithm for face detection and location based on human facial rectangular characteristics in the YIQ and the YCbCr color space has been proposed.The algorithm selects the color space by comparing the average of the input image channels of R,G and B with the average of the channel of B,and segments the skin area with the twodimensional OSTU and integral figure.Then,the skin segmentation figure by median filter based on the acreage of the skin area has been processed.In the end,an optimal face region rectangle with the acreage of the skin area which has been processed by median filter has been constructed,and the rectangle to locate the face region has been used.The experiments show that the algorithm has higher accuracy than the traditional face detection algorithms using single color space.
【Fund】: 江西省教育厅科技项目(GJJ13210);; 江西省光电子与通信电子实验室基金(2011001);; 江西师范大学博士启动基金(4029)资助项目
【CateGory Index】: TP391.41
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved