FACIAL EXPRESSION ANALYSIS SYSTEM USING DEEP LEARNING

Emotion care for human well-being is important for all ages. Automatic emotion recognition plays a crucial role in various fields such as healthcare, human-computer interaction (HCI) and security and defense. Capturing the dynamics of facial expression progression in video is an essential and challenging task for facial expression recognition (FER). In this project, we propose a new low-cost and multi-user framework for emotion care system based on big data analysis for patient feelings, where emotion is detected in terms of facial expression. The system works with deep learning techniques on emotional big data to extract emotional features and recognize six kinds (e.g., angry, disgust, fear, happy, sad, surprise, and neutral) of facial expressions in real-time and offline. In addition, a new dataset for emotion recognition is collected to train the DCNN model. A deep convolutional neural network (DCNN) is further applied to the whole facial observation to learn the global characteristics of six different expressions.And the performed facial expression and predicted results can be saved into device to help cares to observe facial expressions of patients at any time and provide better suggestions to patients..