[1]贺惠新,崔子栋,周逸飞.反射分量分离的特征融合图像翻拍检测[J].华侨大学学报(自然科学版),2022,43(2):222-228.[doi:10.11830/ISSN.1000-5013.202010044]
　HE Huixin,CUI Zidong,ZHOU Yifei.Feature Fusion Diagram Recaptured Detection Based on Reflection Component Separation[J].Journal of Huaqiao University(Natural Science),2022,43(2):222-228.[doi:10.11830/ISSN.1000-5013.202010044]
点击复制

反射分量分离的特征融合图像翻拍检测()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第43卷
期数:
2022年第2期
页码:
222-228
栏目:
出版日期:
2022-03-08

文章信息/Info

Title:
Feature Fusion Diagram Recaptured Detection Based on Reflection Component Separation
文章编号:
1000-5013(2022)02-0222-07
作者:
贺惠新;  崔子栋;  周逸飞
华侨大学 计算机科学与技术学院, 福建 厦门 362021
Author(s):
HE Huixin;  CUI Zidong;  ZHOU Yifei
College of Computer Science and Technology, Huaqiao University, Xiamen 361021, China
关键词:
图像翻拍检测;  同态滤波;  反射分量分离;  特征融合
Keywords:
diagram recaptured detection;  homomorphic filtering;  reflection component separation;  feature fusion
分类号:
TP274
DOI:
10.11830/ISSN.1000-5013.202010044
文献标志码:
A
摘要:
针对翻拍图像对相关人脸识别系统的欺骗性,将基于同态滤波的反射分量分离特征作为新特征加入图像中,形成四通道图像,并将其作为简单卷积神经网络的输入的翻拍图像检测方法.结果表明:在拍摄环境复杂、干扰噪音较多、活体数量较大的环境下,该方法有较高且稳定的准确率、较好的鲁棒性及使用价值.
Abstract:
In the view of the deception of the recaptured image to the relevant face recognition system, the reflection component separation feature based on homomorphic filtering is added to the image as a new feature to form a four channel image, which is used as the detection method of diagram recaptured based on the simple convolution neural network input. The results show that this method has high and stable accuracy, high robustness and application value in the environment of complex shooting case, more interference noise and large number of living bodies.

参考文献/References:

[1] 党永成.人脸识别技术综述及分析[J].电子技术与软件工程,2018(3):158.
[2] 杨巨成,刘娜,房珊珊,等.基于深度学习的人脸识别方法研究综述[J].天津科技大学学报,2016,31(6):1-10.DOI:10.13364/j.issn.1672-6510.20160050.
[3] YANG Taigman,YANG Ming,RANZATO M,et al.Deepface: Closing the gap to human-level performance in face verification[C]//IEEE Conference on Computer Vision and Pattern Recognition.Columbus:IEEE Computer Society,2014:1701-1708.DOI:10.1109/CVPR.2014.220.
[4] SUN Yi,WANG Xiaogang,TANG Xiaoou.Deep learning face representation from predicting 10,000 classes[M].Cham:Springer,2016:1-8.
[5] SCHROFF F,KALENICHENKO D,PHILBIN J.Facenet: A unified embedding for face recognition and clustering[C]//IEEE Conference on Computer Vision and Pattern Recognition.Boston:IEEE Press,2015:815-823.DOI:10.1109/CVPR.2015.7298682.
[6] 夏志强.人脸识别综述[J].电子世界,2017(23):74,76.DOI:10.19353/j.cnki.dzsj.2017.23.033.
[7] CHELLAPPA R,WILSON C L,SIROHEY S.et al.Human and machine recognition of faces: A survey[J].Proceedings of the IEEE,1995,83(5):705-741.DOI:10.1109/5.381842.
[8] 景晨凯,宋涛,庄雷,等.基于深度卷积神经网络的人脸识别技术综述[J].计算机应用与软件,2018,35(1):223-231.DOI:10.3969/j.issn.1000-386x.2018.01.039.
[9] 孙鲲.一种基于LBP水印特征和细粒度识别的图像翻拍检测方法: CN111882525A[P].2020-11-03.
[10] CHEN Rui,LIN Xirong,DING Tianhuai.Liveness detection for iris recognition using multispectral images[J].Pattern Recognition Letters,2012,33(12):1513-1519.DOI:10.1016/j.patrec.2012.04.002.
[11] 谢哲,王让定,严迪群,等.基于同态补偿翻拍图像的方向预测方法[J].计算机应用,2014,34(9):2687-2690,2715.DOI:10.11772/j.issn.1001-9081.2014.09.2687.
[12] 卢燕飞,冯莉,李兴华,等.基于图像表面梯度的翻拍检测[J].北京交通大学学报,2012,36(5):57-61,67.DOI:10.3969/j.issn.1673-0291.2012.05.010.
[13] 钱芸,张英杰.水平集的图像分割方法综述[J].中国图象图形学报,2018,13(1):7-13.
[14] GONZALEZ R C,WOODS R E.Digital image processing[M].2nd ed.Upper Saddle River:Prentice Hall,2002.
[15] 李炜,黄心汉,王敏,等.基于二维高斯曲面拟合的图像灰度补偿算法[J].华中科技大学学报(自然科学版),2004,32(2):43-45.DOI:10.13245/j.hust.2004.02.015.

备注/Memo

备注/Memo:
收稿日期: 2020-10-29
通信作者: 贺惠新(1984-),男,高级工程师,博士,主要从事图像处理、文本数据挖掘的研究.E-mail:hehuixin@hqu.edu.cn.
基金项目: 国家社会科学基金面上资助项目(19BXW110); 华侨大学科研基金资助项目(605-50Y17031)
更新日期/Last Update: 2022-03-20