Emotion component analyzing method and system based on emotion distribution learning

Abstract

The invention discloses an emotion component analyzing method and system based on emotion distribution learning. The method includes marking a basic emotion of each image; calculating a correlation coefficient between emotion distribution vectors and each pair of basic emotion mark vectors and calculating a weight matrix based on the correlation coefficient; by using an image feature vector and emotion distribution thereof as a training set, combining a maximum entropy model with Jeffrey divergence and weight matrix and combining with two regularization items for generating a target function, and optimizing the target function for obtaining a parameter model for forecast of emotion distribution; and performing feature extraction on an image waiting for emotion distribution estimation and using the model obtaining through training for emotion distribution prediction. If the value corresponding to the emotion mark is greater than a constituent ratio of the virtual mark, the emotion is judged to be a main emotion component. By using the method and system provided by the invention, a model for emotion component analysis can be obtained quickly and effectively through training and which emotions are contained in an expression and the proportion of the emotions can be calculated out.

Description

technical field [0001] The invention relates to pattern recognition and machine learning, in particular to emotion distribution learning and face emotion component analysis. Background technique [0002] The facial emotion component analysis is based on a facial expression image, estimating the composition ratio of the six basic emotions in the image. The current main application areas of facial expression analysis technology include: 1) human-computer interaction, the computer gives a series of responses by estimating the emotions contained in facial expressions; 3) In the medical field, expression analysis can be used in robot surgery operations and electronic nurse care, and can detect changes in patients' physical conditions in time according to changes in facial expressions , to avoid tragedy. Therefore, the research and development of facial expression analysis technology is of great significance. In previous studies, facial expression recognition methods often only...

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: Aiming at the problem that the existing expression analysis method only recognizes one emotion of expression, the present invention proposes an emotion component analysis method and system based on emotion distribution learning, which can learn all the emotions and emotions contained in human expression images. Its proportion is more in line with the situation in real life and can better match the application in real life

Images