Human Emotion Detection And Classification Using Convolution Neural Network
European Journal of Molecular & Clinical Medicine,
2020, Volume 7, Issue 6, Pages 237-245
AbstractFacial expression is the non-verbal communication that provides some related- information about the emotion of a person. Detecting and recognizing human emotion is a big challenge in computer vision and artificial intelligence. The main objective of this paper is to develop a robust technique that can detect and identify human emotions such as anger, sorrow, happiness, surprise, fear, disgust and neutral in real-time. In this paper, Viola–John’s algorithm has been proposed to detect the emotions more accurately. This algorithm is used for tracking the emotions in real time. In this paper, the real-time images are captured, and then features are extracted from the face images. Image enhancement and gradient detection are used after extracting the features of face images, then multiple feature extraction like GLCM, LBP and PCA are applied. All extracted features of face images are combined and compared with different databases to get accurate emotion state. CNN classifier is used to achieve conditional detection of accurate facial expressions.
- Article View: 177
- PDF Download: 550