Facial expression recognition method based on feature extraction

Abstract

The invention discloses a facial expression recognition method based on feature extraction, which comprises the following steps: firstly, dividing a facial expression image into sub-regions, adopting two AR-LGBP operators with different sizes for the same pixel point to obtain two eight-bit binary sequences, carrying out logic exclusive-or operation on the two binary sequences in one-to-one correspondence to obtain a new binary sequence, and carrying out feature extraction on the new binary sequence; wherein a decimal numerical value converted from the sequence is the pixel value of the pixel point, calculating the pixel value of each pixel point in each sub-region according to the method to obtain a histogram of the sub-region, and connecting the histograms of the sub-regions to generate a facial expression feature vector; and finally, performing dimensionality reduction on the generated feature vector through a principal component analysis algorithm, and performing facial expression classification recognition in combination with an SVM classifier. According to the method, the pixel relation between neighborhoods is considered, the feature description capability can be improved, expansibility is achieved, and features can be extracted under different scales.

Description

technical field [0001] The invention relates to the technical field of digital image processing, in particular to a feature extraction technology with high efficiency, low complexity, good robustness and strong discrimination ability adopted in the expression recognition process. Background technique [0002] At present, the research on artificial intelligence has reached a high level, while the research on human emotion and cognition is relatively small. In real life, people expect computers to serve the society like human beings, and to be more intelligent in human-computer interaction, but it is far from enough to have the perception capabilities of vision and hearing, and it is necessary to add emotional understanding and emotional recognition functions . The visual information reflected by the human face is the most direct and important carrier of human emotional expression and interaction. Researchers can guess the true inner thoughts of the expresser through the chan...

AI Technical Summary

Problems solved by technology

LGBP has a stronger discriminative ability than LBP, but the operator is susceptible to noise, and the size of the neighborhood is fixed, so it cannot extract texture features well at large scales.

Images