A Facial Expression Recognition Method Based on AlexNet Deep Learning Algorithm

Facial expression recognition is crucial in various domains, including healthcare, affective computing, traffic safety, and human-computer interaction. However, traditional facial expression recognition methods based on handcrafted features often suffer from limitations such as low accuracy, poor generalization, and high computational overhead. The paper proposes a facial expression recognition method based on the AlexNet deep learning architecture, enhanced with data augmentation techniques to improve robustness and generalization. Compared to ResNet, the proposed method achieves a superior trade-off between recognition accuracy and computational efficiency. Experimental results on the Emotion-Domestic dataset demonstrate that the AlexNet-based algorithm achieves a recognition accuracy of 92%. The results validate the effectiveness of the proposed approach as a computationally efficient and accurate solution for real-time facial expression recognition.