Expression recognition method based on brain-computer collaborative intelligence

Abstract

The invention relates to an expression recognition method based on brain-computer collaborative intelligence. The method mainly adopts a two-layer convolutional neural network to extract image visualfeatures of human face expressions, adopts a plurality of gating circulation units to extract electroencephalogram emotion features induced when the expressions are watched, establishes a mapping relation between the two features through a random forest regression model, and finally adopts K-neighbor classifier to carry out expression classification on the predicted electroencephalogram emotion features obtained by the regression model. The method comprises the steps of data acquisition, data preprocessing, image visual feature extraction, electroencephalogram emotion feature extraction, feature mapping and expression classification. Expression classification results show that a good classification result is obtained by adopting the predicted electroencephalogram emotion features. Comparedwith a traditional image vision method, the expression recognition method based on brain-computer collaborative intelligence is a promising emotion calculation method.

Description

technical field [0001] The invention belongs to the field of expression recognition in the field of emotion computing, and in particular relates to an expression recognition method based on brain-computer collaborative intelligence. Background technique [0002] In general, we use facial expressions, body posture, and voice pitch to infer someone's emotional state (such as joy, sadness, and anger, etc.). As far as carrying emotional meaning, facial expression is the main source of information in daily communication, and it is also a key component of human-computer interaction system (HCIS). How to effectively recognize facial expressions has become an important topic. [0003] At present, facial expression recognition (FER) methods mainly start from the perspective of human vision, and use computer technology to extract relevant features from facial images to distinguish different expressions. According to whether the features are artificially designed or generated by neural...

AI Technical Summary

Problems solved by technology

However, the multi-layer network training time will increase, and the training difficulty will increase accordingly.

At the same time, the requirements for the size and diversity of training samples will also become higher, otherwise the overfitting of the model will also lead to unsatisfactory recognition results

Images