欢迎光临,现在是     设为首页   加入收藏   联系我们
文章摘要
肖 童,李京兵,曹春杰.基于双混沌映射与神经网络的加密人脸识别鲁棒性算法研究[J].海南大学学报编辑部:自然科学版,2018,36(4):.
基于双混沌映射与神经网络的加密人脸识别鲁棒性算法研究
A Robust Algorithm of Encrypted Face Recognition Based on double chaos map and Neural Network
投稿时间:2018-06-03  修订日期:2018-08-29
DOI:
中文关键词: 人脸识别; 双混沌映射; 神经网络; 鲁棒性
英文关键词: Face recognition; double chaos map; Neural Network; Robustness
基金项目:国家自然科学基金(61263033), 海南省重点研发计划(ZDYF2018129)
作者单位E-mail
肖 童 海南大学 信息科学技术学院 xiaotong920923@sina.com 
李京兵 海南大学 信息科学技术学院 jingbingli2008@hotmail.com 
曹春杰 海南大学 信息科学技术学院  
摘要点击次数: 98
全文下载次数: 22
中文摘要:
      人脸识别在监控,金融,信息安全等领域具有广泛应用,但用户信息可能会遭到窃取而泄露,针对传统人脸识别的安全性问题提出了一种基于双混沌映射与神经网络的加密人脸识别鲁棒算法。首先利用Logistic与Henon混沌映射产生密钥,然后利用密钥在DWT-DCT变换域内加密人脸图像,结合PCA与神经网络实现对加密人脸的识别,最后进行常规攻击,几何攻击,遮挡攻击来测试其鲁棒性。实验结果表明,该算法具有良好的加密效果与鲁棒性。
英文摘要:
      Face recognition application has been in the areas of surveillance, finance, and information security. In view of the security problems of the traditional face recognition method, this paper proposes a robust algorithm of encrypted face recognition based on double chaos map and neural network. Firstly, we used the Logistic Map and Henon Map to generated the key, and then encrypt face image in DWT-DCT transform domain, combined with PCA and neural network to recognize the encrypted face. Finally, we test the robustness by conventional attack and geometric attack and occlusion attack. Experimental results show that the algorithm has good encryption effect and it has strong robustness.
关闭