Real-Time Emotion Recognition Using Convolutional Neural Network: A Raspberry Pi Architecture Approach

Facial identification of emotions is an area of study that has several cutting-edge solutions, aimed at applications in areas such as security, marketing and robotics. In the literature, there are numerous articles that present algorithms from various perspectives to perform this task to improve the understanding of human behavior and improve efficiency in the performance of tasks in the future. Emotions can manifest themselves in a variety of ways, such as textual, vocal, verbal, and facial expressions. It is particularly crucial to pay attention to facial expressions as they provide useful information for social interaction and proper communication by observing a person’s actions. This paper presents an architecture that could identify and classify seven facial emotions in children. To achieve this, the Raspberry Pi 3b+ development board was used, and real-time feature extraction is performed using a Convolutional Neural Network (CNN). The process consists of three main stages: face detection, facial feature extraction and facial emotion classification using the FER-2013 dataset. The results obtained for facial emotion recognition show a satisfactory identification rate of 95%, which demonstrates that the use of this architecture is feasible to implement in future robotics projects.