A Study on Emotion Classification of Image Based on BP Neural Network

摘要

In this paper, we adopt BP neural network as classification method and established the emotion model to simulate people's emotion based on image template. For achieving this purpose, firstly, we used 9 images which has different features of the famous psychologist Furnham’s Shape & Color Test as templates and the basis of emotion classification. Then, we got a vector which has 27 dimensions by using part-division method on color character and normalized them in the HSV color space, used Hu moment invariants to described image template shape feature. At last, we carry out classification experiments with BP neural network model. The experimental results show that BP neural network model can achieve classification of personalized emotion based on image template.