SignsWorld Facial Expression Recognition System (FERS)

Live facial expression recognition is an effective and essential research area in human computer interaction (HCI), and the automatic sign language recognition (ASLR) fields. This paper presents a fully automatic facial expression and direction of sight recognition system, that we called SignsWorld Facial Expression Recognition System (FERS). The SignsWorld FERS is divided into three main components: Face detection that is robust to occlusion, key facial features points extraction and facial expression with direction of sight recognition. We present a powerful multi-detector technique to localize the key facial feature points so that contours of the facial components such as the eyes, nostrils, chin, and mouth are sampled. Based on the extracted 66 facial features points, 20 geometric formulas (GFs), 15 ratios (Rs) are calculated, and the classifier based on rule-based reasoning approach are then formed for both of the gaze direction and the facial expression (Normal, Smiling, Sadness or Surprising). SignsWorld FERS is the person independent facial expression and achieved a recognition rate of 97%.