Scholarlink

<p>In academic research, the study of recognition of facial emotions is of considerable importance as there are several implications related to this in scenarios related to the assessment of mental health as well as human–computer interaction. The presented work deals with a novel approach to recognizing facial gestures by integrating an accelerated KAZE (AKAZE) algorithm and deep Convolutional Neural Network (CNN). The most prominent textural features of facial images can be extracted by employing the AKAZE algorithm and the CNNs can be utilized for acquiring the informative representations to classify the emotions in a precise manner. Benchmark datasets can be used for evaluating the presented methodology. The obtained results demonstrate the efficacy of the presented approach in comparison to existing methods in terms of accurate identification of expressions of facial emotions. By combining the AKAZE algorithm and CNNs, our method shows improved accuracy and resilience, with the potential to significantly enhance the evaluation of mental health, thus enabling more adaptive human–computer interaction.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Hybrid Approach to Facial Emotion Recognition: Integrating AKAZE Algorithm and Deep Learning

  • 期刊论文

  • 发表日期：

    2025年5月7日

查看全文
  • Navneet Kaur,
  • Kanwarpreet Kaur

摘要

In academic research, the study of recognition of facial emotions is of considerable importance as there are several implications related to this in scenarios related to the assessment of mental health as well as human–computer interaction. The presented work deals with a novel approach to recognizing facial gestures by integrating an accelerated KAZE (AKAZE) algorithm and deep Convolutional Neural Network (CNN). The most prominent textural features of facial images can be extracted by employing the AKAZE algorithm and the CNNs can be utilized for acquiring the informative representations to classify the emotions in a precise manner. Benchmark datasets can be used for evaluating the presented methodology. The obtained results demonstrate the efficacy of the presented approach in comparison to existing methods in terms of accurate identification of expressions of facial emotions. By combining the AKAZE algorithm and CNNs, our method shows improved accuracy and resilience, with the potential to significantly enhance the evaluation of mental health, thus enabling more adaptive human–computer interaction.