Year: 2025
Type: article
Abstract: Facial expression recognition is an important research direction in the field of computer vision, which has a wide range of application potential, including human-computer interaction, emotional computing, security monitoring, and so on. In this study, a facial expression recognition method based on the TensorFlow framework is proposed, which uses a convolutional neural network (CNN) to automatically extract facial features and classify emotions. By training with the FER2013 dataset, we construct a multi-layer convolutional neural network model and use data enhancement technology to improve the generalization ability of the model. Experimental results show that the proposed method can effectively identify seven basic emotions (happiness, sadness, anger, disgust, surprise, fear, and neutrality), and the classification accuracy on the FER2013 dataset reaches 70%, which is superior to other traditional facial expression recognition methods. Through the analysis of the confusion matrix and classification report, we find that the model is confused in some emotional categories (such as fear and sadness), and the accuracy and robustness can be further improved by optimizing the model structure or introducing stronger regularization methods in future work.
Author: Jiyang Gao
Language: English
Topic: Emotion and Mood Recognition
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4409238763