Emotion classification on image using Hidden Markov Models

The area of information technology related to the recognition of emotions in images of people being actively developed. Emotion recognition by Hidden Markov Models using the image of a human face is presented in this paper. As the sequence of observing states the distance used between the parts of the face (the distance between the corners of the lips and eye line, between the corners of the eyebrows and eyes etc.), and the size of the face (width of mouth etc.). We consider two different approaches to the classification of the image of emotions using Hidden Markov Models. The first method is based on determination of the maximum of the likelihood function. The second approach is based on multi-level Hidden Markov Model. Also results which have been calculated for the couple different emotions are presented at the end of the article.