Facial Expression Recognition Using Adaptive Neuro-fuzzy Inference Systems
Isidoros Perikos, Michael Paraskevas, Ioannis Hatzilygeroudis · 2018
Facial expressions form a universal language of emotions, which can instantly express a wide range of emotional states and feelings. The analysis of facial expressions and the accurate recognition of their emotional content are highly desired and assistive in a wide spectrum of domains. In this paper, we present a work on the analysis of facial expressions and the recognition of emotions using an approach that is based on adaptive neuro fuzzy inference systems. Initially, given a new image, human faces are detected using the Viola-Jones algorithm. Then, facial expressions are analyzed and facial deformations of specific regions such as eyes, eyebrows and mouth are located and then characteristics such as locations, length, width and shape are extracted. The feature vectors represent the deformations of the facial expression and based on them the emotional content of the facial expressions is recognized using an approach that is based on adaptive neuro fuzzy inference systems. An evaluation study was conducted on the JAFFE database and the results collected were very encouraging indicating that the approach is efficient and accurate in analyzing facial expressions and recognizing their emotional content.