摘要
人脸识别技术是生物特征识别的一个重要分支,但在人脸识别过程中往往存在稳定性差、不易识别等问题,这些问题大多发生在识别的检测环节.基于黄金分割对人脸的眉、眼、耳、鼻、口五官进行划分,抓取主要特征,提出基于黄金分割的人脸检测算法,并通过Biometrics Ideal Test人脸数据库和FDDB人脸数据库进行对比试验.结果表明:基于该算法构建的检测框架稳定性好,对不同角度或inthewilds的人脸有较好的检测性能和较快的检测速率.
Face recognition technology is an important branch of the biometric identification,but there are some problems such as poor stability and difficult to identify in the process of face recognition,most of these problems occur in the detection process of identification.Based on the golden section,the facial main features of face,eye,ear,nose and mouth are divided and captured,the face detection algorithm based on golden section is proposed and contrasted the tests though Biometrics Ideal Test face database and FDDB face database.The results indicate that the detection framework based on proposed algorithm has better stability and detection performance and faster detection rate for the faces of different angles or in the wilds.
作者
王茹
贺兴时
Wang Ru;He Xingshi(College of Science,Xi’an Polytechnic University,Xi’an 710048,China)
出处
《北华大学学报（自然科学版）》
CAS
2019年第4期556-560,共5页
Journal of Beihua University（Natural Science）
基金
陕西省教育厅科研计划项目(18JK0333)
关键词
人脸识别
人脸检测
黄金分割
face recognition
face detection
golden section
