计算机应用与软件2017，Vol.34Issue(2)：203-208,213,7.DOI:10.3969/j.issn.1000-386x.2017.02.036
一种基于多特征融合的二维人脸欺诈检测方法
2D FACE SPOOFING DETECTION METHOD BASED ON MULTI-FEATURE FUSION
摘要
Abstract
To address the issue that current face biometric systems are vulnerable to spoofing attacks,a novel face spoofing detecting approach based on the analysis of facial images is proposed.The proposed method absorbs the advantages of LBP,Gabor wavelet and pixel texture feature.A nonlinear SVM classifier is used to distinguish the human faces and face prints.Compared with the previous works,the proposed method does not require user cooperation and is has high availability.In addition,the proposed fusion texture feature can also be used for face recognition,which provides a unique feature space to deal with spoofing detection and face recognition.The experimental results show that this method is better than current methods and has higher recognition accuracy.关键词
人脸欺诈检测/纹理分析/多特征融合/LBP/Gabor小波/像素特征Key words
Face spoofing detection/Texture analysis/Multi-feature fusion/LBP/Gabor wavelet/Pixel Feature分类
信息技术与安全科学引用本文复制引用
袁海聪,李松斌,邓浩江..一种基于多特征融合的二维人脸欺诈检测方法[J].计算机应用与软件,2017,34(2):203-208,213,7.基金项目
海南省重大科技项目(JDJS2013006) （JDJS2013006）
海南省应用技术开发项目(ZDXM2015103). （ZDXM2015103）
