作者: M Karthigayan , M Rizon , Sazali Yaacob , R Nagarajan , M Sugisaka
DOI: 10.1109/ICCAS.2007.4406868
关键词: Feature extraction 、 Face (geometry) 、 Contextual image classification 、 Computer science 、 Fuzzy set 、 Facial recognition system 、 Artificial intelligence 、 Pattern recognition 、 Fitness function 、 Fuzzy clustering 、 Ellipse 、 Computer vision
摘要: In this paper, lip and eye features are applied to classify the human emotion using a set of irregular regular ellipse fitting equations genetic algorithm (GA). A South East Asian face is considered in study. The parameters relating emotions, either case, entirely different. All six universally accepted emotions one neutral for classifications. method which fastest extracting adopted Observation various subject lead unique characteristic lips eyes. GA optimize characteristics each emotion. That is, top portion configuration part bottom different ellipse. Two based fitness proposed relevant that define listed. One function configuration. has achieved reasonably successful classification some classification, optimized data values messed or overlapped other ranges. order overcome overlapping problem between at same time improve fuzzy clustering (FCM) approach been implemented offer better classification. GA-FCM offers good within ranges clusters.