Year: 2025
Type: article
Abstract: ABSTRACT Advanced facial expression recognition technology can significantly enhance human‐computer interaction and improve intelligent services for humans. This paper introduces a novel facial expression recognition method utilizing an enhanced ConvNeXt network. By integrating the SENET attention mechanism into the ConvNeXt block, key feature information extraction is effectively enhanced. Additionally, the incorporation of the focal loss (FL) function optimizes the classification performance of the network model. Experimental results show that the improved ConvNeXt network achieves higher accuracy compared to other deep learning models, with accuracy rates of 83.8% and 70.4% on the RAF‐DB and FER2013 datasets, respectively.
Source: IET Image Processing
DOI: 10.1049/ipr2.70118
Institutions: Liaoning Shihua University, Shenyang Agricultural University
Language: English
Topic: Emotion and Mood Recognition
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $2,740
Award: G1369544492
OpenAlex ID: W4410901342