A face image recognition algorithm based on DFT encryption domain

Abstract

Information security and privacy protection are increasingly becoming the mainstream topics nowadays. While information is transmitted to third party, how to ensure that the transmitted information is not leaked has been a research focus. While most face recognition algorithms are in progress in the plaintext domain, how to solve the problem of encrypted face recognition in the cloud computing environment is of significant importance. Based on Discrete Fourier Transform (DFT) and Logistic chaotic map, we proposed a face image recognition algorithm based on DFT encrypted domain. We encrypt the image in frequency domain. The DFT is performed on the encrypted face image to extract the feature vector and establish a feature database. The feature vector functions as the “Keyword” of face recognition and matches the similarity by calculating the NC (Normalized Cross Correlation Coefficient, NC) between its feature vector and the feature vector of encrypted face image stored in the feature database. Finally, through finding out the corresponding encrypted image with the greatest value of NC we realize the face recognition. The experimental results demonstrate that this algorithm has ideal ability to resist the conventional attack, geometric attack, illumination attack and mask attack.