Facial Emotions Recognition System Using Hybrid Convolutional Neural Network Model

This work presents a real-time facial emotion detection method for detecting human emotions. The most significant way to depict human emotions is through facial expressions, which are also important research topics for computer vision and artificial intelligence systems. The primary feature of the human body that is utilized for nonverbal communication is the face. This paper proposed a workable working model to identify seven human emotions on a single face and a group of faces. Image quality, obscuring variations of hidden body position, occlusion, and variable lighting conditions make it difficult to identify a group of emotions. Crowd analysis, social media, signage, social event detection, public safety, human–computer interaction, and many other fields all use the emotion detection group domains. Three detection cycles make up the suggested enhanced hybrid convolutional neural network model-long short-term memory (IHMCNN-LSTM) facial emotion detection system: face detection, facial feature detection, and emotion detection. Real-time images and videos are captured by the web camera and then processed through the HAAR cascade method for human face detection.