Recognition algorithm of face recognition

Generally speaking, face recognition system includes image acquisition, face location, image preprocessing and face recognition (identity confirmation or identity search). The input of the system is generally one or a series of face images with undetermined identities and several face images with known identities or corresponding codes in the face database, while the output is a series of similarity scores, indicating the identities of the faces to be recognized.

Classification of face recognition algorithms

Feature recognition algorithm based on facial feature points.

Recognition algorithm based on whole face image.

Recognition algorithm based on template.

Recognition algorithm using neural network.

Theory based on illumination estimation model

An illumination preprocessing method based on gray correction is proposed, and corresponding illumination compensation and illumination balance strategies are carried out on the basis of illumination estimation model.

Optimal deformation statistical correction theory

Based on statistical deformation correction theory, face posture is optimized; Reinforcement iteration theory

Strengthening iterative theory is an effective extension of DLFA face detection algorithm.

Original real-time feature recognition theory

This theory focuses on the median processing of real-time face data, so that the recognition speed and efficiency can achieve the best matching effect.