Security algorithm of face recognition based on local binary pattern and random projection

With the growing applications of biometrics, there are many concerns about the security and privacy of the biometrics data. This paper proposes a security algorithm of face recognition, which bases on the local binary pattern (LBP) and random projection. Feature extraction is first performed on face images by LBP operator. Random projection is done using the random matrix which is generated by user specific key. All the above successive procedures produce the cancelable and non-invertible feature data, which is stored in template databases and can achieve the protection of biometrics data. In extensive experiments with publicly available face datasets ORL, YALE and FERET, higher recognition accuracy is reached, which demonstrates that the proposed approach is able to not only protect the biometrics data, but also is robust to various complex conditions, such as illumination, changes in the pose and expression, etc.