Extraction method for facial expression feature

Abstract

The invention provides an extraction method for facial expression feature. The method comprises the steps that a facial expression image is divided into N blocks of images, and the size of each branch image is m * n; the coded value of LWBP 1 and LWBP 2 of all pixels in each image are calculated by a locally weighted binary pattern namely LWBP; the LWBP column diagram of each image is counted; two column diagrams of each image are directly overlapped to get a column diagram to serve as the final LWBP feature of each image; all the statistical histograms of all blocks of the images are connected in sequence to obtain a LWBP feature vector, used for classification and identification, of the whole image. By calculating weighting gray values of two-group symmetrical eight templates respectively, the size of the weighted values and equal weighted values in every direction are compared and coding is conduced, the gray level change of neighborhood pixels in different directions are considered comprehensively, the detail feature of the facial expression can be effectively represented, certain robustness is possessed for noise, the identification speed is quickened obviously, and the practicality is possessed.

Description

technical field [0001] The invention relates to a method for extracting human facial expression features. Background technique [0002] Facial expression contains rich human behavior information, is a form of expression of human emotions, and is also an effective and important means for people to carry out non-verbal communication. People can accurately, fully and subtly express their thoughts and feelings through facial expressions, and can also identify each other's attitude and inner world through facial expressions. Therefore, research on expression recognition has important academic value and application prospect, and has gradually become a research hotspot in recent years. [0003] Facial expression recognition is the process of computer feature extraction and classification of facial expression information, which enables computers to infer human psychology from human expressions, thereby realizing advanced intelligent interaction between humans and computers. The fa...

AI Technical Summary

Problems solved by technology

This encoding method is easily affected by changes in the gray level of the neighborhood and is sensitive to noise.

[0006] 2. The LBP algorithm performs 8-bit encoding on each block (block) image, and the obtained feature dimension is the number of blocks (block) × 28, resulting in an excessively large image feature dimension, which reduces the recognition speed and also affects the recognition rate. , which is more obvious on large databases

Images