Facial expression recognition method with noise robust

Abstract

The invention discloses a facial expression recognition method with noise robust. The method comprises steps of improvement on filter pre-treatment and improvement on a feature extraction operator. According to each anisotropic diffusion filtering model of a relative brightness difference adjustment factor, defects that each original anisotropic diffusion filtering method smoothes out details while noise is filtered can be remedied, facial noise and weak detail expression information are distinguished, and integration of expression image information can be ensured; and in addition, the improved HOG operator is adopted to extract expression features, the feature vector dimension is reduced, interference of redundant information is reduced, the algorithm operating time is shortened, classification precision and noise robust are improved, and good application prospect is provided.

Description

technical field [0001] The invention relates to a noise-robust human facial expression recognition method, which belongs to the technical field of human face recognition. Background technique [0002] Facial expression is one of the important body languages ​​of human beings, which can accurately reflect changes in emotional, mental, and psychological states. In recent years, the use of computers to analyze and understand facial expressions to complete related work has important application prospects in human-computer interaction. Facial expression recognition technology has gradually become a research hotspot. The facial expression recognition system mainly includes image preprocessing, face detection and area Segmentation, expression feature extraction and classification are four parts. Considering that the recognition effect of the classifier largely depends on the accuracy of the feature description, the expression feature extraction is an important part of the facial ex...

AI Technical Summary

Problems solved by technology

[0011] The purpose of the present invention is to overcome the problem of traditional facial expression recognition. The noise-robust human facial expression recognition method provided by the present invention is based on the anisotropic diffusion filter model of the relative brightness difference adjustment factor, which compensates for the original anisotropic The heterogeneous diffusion filtering method can smooth out the lack of details while filtering out noise, distinguish facial noise and weak detail expression information, and ensure the integrity of expression image information. At the same time, the improved HOG operator is used to extract expression features and reduce the dimension of feature vectors. Reduce the interference of redundant information, shorten the running time of the algorithm, improve the classification accuracy and robustness to noise, and have good application prospects

Images