Improving the Classification of Emotions by Studying Facial Feature
Yahia Siwar, E. Ridha, J. Olfa, Z. Mourad · International Journal of Computer Theory and Engineering · 2015
This work aims to recognize the six basic emotions using facial expression and improve the classification in term of time and space memory.We started with the idea that emotions can be absolutely distinctive and from a single feature we can recognize emotion and therefore we save time in learning data.We also noted that the similarity between the characteristics of two emotions can cause a lot of errors.So we concluded that it is necessary to specify the characteristics for each emotion to improve classification.The study of various facial features (eyes, eyebrows and mouth) of each emotion has helped us to find a new use of face features and ultimately lead to a faster classification.