.png){kind=icon}
.png){kind=icon}
Return
Efficient blind face recognition in the cloud
.png){kind=icon}
PRE
AI
DOI：10.1007/s11042-019-08280-y.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Nowadays, with the maturity and wide application of face recognition technology, the recognition accuracy, recognition efficiency, and data security have attracted people's attention. However, when face recognition is performed, face information is completely exposed to the cloud server without any protection measures. Therefore, a series of problems caused by insecure face information is coming. Can we find a way to prevent uncontrolled use of facial information without cloud protection and improve recognition efficiency and accuracy? Given this situation, we have proposed two options. The first requires a third-party library; the second does not require a third-party library. The first scheme is efficient privacy preserving face identification in the cloud through sparse representation, which relies on the third-party face image database, and the first scheme is simply referred to as SRBased. The second scheme is efficient privacy preserving face identification in the cloud based on deep neural network, which does not depend on the third-party face image database, and the second scheme is simply referred to as DNNBased. Both schemes can be divided into two parts: client and cloud server. The client is responsible for acquiring face images, and the server is responsible for recognizing and calculating. Through homomorphic encryption and OT protocol, secure face recognition is realized. In the whole recognition process, the server does not need to decrypt the image data. In the two schemes, the client and the server will not get any information from each other. Even if the third party intercepts the ciphertext in the transmission process, it will not get any information under the premise of private key security. Therefore, the two schemes can achieve the purpose of protecting privacy and security. The experimental results show that the efficiency of the two schemes is greatly improved compared with SCiFI schemes. The second scheme improves recognition accuracy greatly.
Keywords:
Privacy protection
Face recognition
Sparse representation
Cloud computing
Homomorphic encryption
Deep neural network
Deep learning
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W