Method for classification human facial expression and semantics judgement quantization method

Abstract

The invention relates to in an analysis method of the Facial Expression image classification and semantic evaluation on the basis of semantic evaluation and the quantization. This invention makes use of the sample collection of training Facial Expression images, extracting expression characteristics to form a marker map of LG Vector, projecting it to the Principal Component PCA subspace, making the use of these reduced-order LG Vectors to learn mixed multi-dimensional t-distribution, as the semantic judgment of six basic emotions of the images, in according to that which probability taken up by expression is maximal, that image is judged thereby to that expression. Which resolve the recognition difficulties of the uncertainties, automatic facial expression existed in the current technology, and the singularity difficult to overcome such defect. This invention is an extremely flexible and powerful modeling tools based on the statistics, provides a more robust ways, avoiding the extreme estimates of posterior probability of the subordinate component of sample observed, the training samples do not require labeling, do not require any post-processing, less sensitive to outliers, avoid this artificial Judgment.

Description

technical field [0001] The invention relates to an analysis method of facial expression images, in particular to an analysis method for classifying and quantifying the facial expression images on the basis of semantic evaluation. Background technique [0002] At present, research on facial expression recognition from the perspective of computer vision and pattern recognition is a new direction of research on facial expression recognition. The ability to recognize facial expressions is one of the signs of intelligent computers. Human-computer interaction becomes possible. When a computer has the ability to recognize emotions, it has the same ability to recognize the emotions of others as humans, which has broad application prospects in education, medical care, security inspection, entertainment, business, social development and other fields. In addition, through computer analysis and processing methods, the study of emotion is raised from the perspective of perceptual cognit...

AI Technical Summary

Problems solved by technology

However, these six emotion categories are descriptive and, therefore, not very clear-cut; second, it is possible for an expression to be assigned to multiple emotion categories, for example, a smiling mouth and a raised eyebrow are a mixture of surprise and delight

These all make the current automatic facial expression recognition method difficult, and the above two methods cannot give a quantitative judgment of facial expression semantics

Although from the perspective of data regression analysis, it is possible to estimate the corresponding basic emotional components of face images, such as the Kernel Canonical Correlation Analysis (KCCA) method, which establishes the relationship between a face image and its corresponding basic emotional data. and then use this relationship model to predict and estimate unknown facial expression images, but this method requires manual semantic evaluation of training samples, and there are still insurmountable singularity problems in regression

Images