Face Database Security Information Verification Based on Recognition Technology

Shumin Xue

《International Journal of Network Security》 21卷4期 (2019/07) Pp. 601-606

https://doi.org/10.6633/IJNS.201907_21(4).09

摘要

In recent years, with the rapid development of the Internet, information can be transmitted more and more rapidly, and the issue of confidentiality and security has become increasingly important. Compared with traditional information verification methods, biometric identification is more secure and convenient. This study used the MATLAB software to carry out the simulation of information verification performance of face recognition algorithm based on Local Directional Pattern Algorithm (LDP) and Principal Component Analysis (PCA). The face image data were from ORL database. The results showed that the increase of the training set samples could raise the accuracy of security information verification of the two algorithms and took less time, and under the same number of training samples, the algorithm of face recognition based on PCA, compared with face recognition algorithm based on LDP, had higher accuracy and less time consuming. In conclusion, PCA-based face recognition algorithm is more suitable for security information verification.

關鍵字

Face Recognition ； Local Directional Pattern ； ORL Face Database ； Principal Component Analysis