Face recognition model based on privacy protection and random forest algorithm

Abstract

With the rapid development of face recognition technology, people's awareness of privacy protection is getting higher and higher. Arnold transform because of its simple and easy to be applied to digital watermarking, security and other fields, has achieved good results. In this paper, Arnold transform is applied to face images, and propose scrambled face recognition of random forest (SFR-RF) model based on random forest (RF) algorithm. The model extracts the features of face database, makes label classification data, forms training set and testing set, and recognizes scrambled face by RF classification algorithm. The experimental results show that the SFR-RF model achieves good results in the area of face image scrambling. It improves the recognition rate by 20% compared with the traditional decision tree algorithm, making the proposed method a good candidate for face privacy protection and identification.