A Robust Algorithm of Encrypted Face Recognition Based on DWT-DCT and Tent Map

摘要

With the progress of the Internet and the development of science and technology, face recognition technology is becoming more and more mature, has been paid more and more attention and application in the fields of monitoring, finance and information security. Compare with the traditional authentication methods, face recognition technology is more stable. However, the information data of face recognition system is still stored in the Internet, and may be attacked by hackers, resulting in the disclosure of personal privacy and data security issues. Therefore, this paper proposes a robust algorithm of encrypted face recognition based on DWT (Discrete Wavelet Transform) and DCT (Discrete Cosine Transform) and Tent Map. Firstly, by using Tent Map we generate pseudo-random sequences and encrypt face image in DWT-DCT domain. Then, combine PCA (Principal Component Analysis) and BP neural network to realize face recognition. Finally, we test the robustness by conventional attack and geometric attack and occlusion attack. The results of simulation experiment on ORL face database show that the recognition rate of this algorithm is higher than of unencrypted algorithm, at the same time, this algorithm can effectively protect the personal information security through encryption, and it has strong robustness and has wide application prospects.