Nowadays, Facial Expression Recognition (FER) or Emotion Recognition (ER) is significant for Security Systems,
Human-Computer Interaction, Lie Detection, Monitoring at ATMs, etc. It’s a system that recognizes the seven basic human
emotions (Happy, Angry, Sad, Fear, Disgust, Surprise and Neutral). Real-Time Emotion Recognition implementation may
become complex in Image Classification. Use of Deep Learning is important in Image Classification. In Deep Learning,
Convolutional Neural Networks (CNNs) can help to reduce difficulties of Emotion Recognition. But Traditional CNN-based
methods, i.e. Shallow CNN (SHCNN), leads to less accuracy of the CNN model. In this paper, we propose a Deep Convolutional
Neural Network (DCNN) architecture to extract features of ER tasks.
