An emotion prediction system based on a deep convolutional network

Abstract

The invention discloses an emotion prediction system based on a deep convolutional network. Further, a further approach is provided, The method comprises the steps of manufacturing a human face emotion sample image library, wherein a convolutional neural network is utilized to train and generate a classifier, the classifier is utilized to carry out a face emotion pre-judgment step, and the step ofmanufacturing the face emotion sample image library comprises the following steps: S11, sample collection: collecting and sorting face expression images; and S12, sample classification: carrying outclassification processing on the collected face expression images. Face emotions are divided into three categories: neutrality, positive and negative. The method is reasonable in design and practicalin function, the original image is directly subjected to feature extraction and then identified, and the identification effect is poor. In order to solve the problem, the number of layers of the neural network is increased when the neural network is built, so that the network can extract deeper features; Wherein the convolutional neural network comprises 24 convolutional layers and 2 full connection layers. The classification effect and recognition robustness of the multi-level convolutional neural network are improved well, and the multi-level convolutional neural network is diversified in structure and suitable for popularization.

Description

technical field [0001] The invention relates to the technical field of facial emotion recognition and image processing equipment, in particular to an emotion prediction system based on a deep convolutional network. Background technique [0002] With the rapid development of computer technology, the interaction between human and computer has become an indispensable part of people's daily life. Traditional human-computer interaction is simply to complete command execution with the help of keyboard, mouse and other tools. With the rapid development of artificial intelligence and pattern recognition technology, biometric technology has become one of the important fields of human-computer interaction research. As a special biological feature, expression is an important means of communication and interaction between people, and it is also an objective indicator for understanding the subjective psychological state of others. In human-to-human communication, only 7% of the informa...

AI Technical Summary

Problems solved by technology

The features extracted by these traditional feature extraction algorithms are greatly affected by the external environment, which makes the final recognition rate not high.

Images