The influence of image enhancement algorithm on face recognition system
Yuxuan He · 2021
Aiming at the low efficiency of face recognition devices such as access control systems due to resolution and uneven illumination in dark environments, the Harr feature is used to detect the face, and the EigenFace method is used to extract the characteristic face information, and a face recognition system is established for simulation the actual situation. Using image enhancement methods such as histogram equalization, Retinex, and Median filtering to process portraits, experiments were conducted on the ORL library and self-collected face training set. The experimental results show that some image enhancement algorithms can significantly improve the face recognition rate, and Image enhancement algorithms are conducive to machine recognition in the field of portraits.