An Appearance Based Method for Emotion Detection from Face Image

Abstract: Humans can use vision to identify objects quickly and accurately through their Emotion. Computer vision trying to emulate human Expression from Digital Image Inputs. In this paper an appearance based method is proposed for emotion detection using face image. In this appearance method DWT (Discrete Wavelet Transform) and PCA (Principle Component Analysis) is applied for feature extraction and SVM (support vector machine) is used for emotion classification. The experiments are conducted on a well-known face image datasets (FERET and Indian Face dataset) and achieved acceptable classification rate. Keywords: Image Pre-processing, Enhancement, Segmentation, Classification, Feature Extraction