Detection of human emotions through facial expressions using hybrid convolutional neural network-recurrent neural network algorithm

• DOI: 10.1016/j.iswa.2024.200339
• Publication Date: 2024-02-09T17:46:15Z
• Authors: Haposan Vincentiu...; Achmad Pratama Rifai
• Keywords: Connectionism

Cognitive science plays a pivotal role in deciphering human behavior by understanding and interpreting emotions prevalent everyday life. These manifest through various cues, including speech patterns, body language, notably, facial expressions. Human expressions serve as fundamental mode of communication interaction. Within the realm computer vision, Facial Expression Recognition (FER) stands crucial field, offering diverse techniques to decode from This research aims develop hybrid Convolutional Neural Network – Recurrent (CNN-RNN) model adept at detecting derived based on video data. The models are developed Emotional Wearable Dataset 2020. dataset consists several expressions, four them - amusement, enthusiasm, awe, liking has never been explored previous datasets. expansion provides more comprehensive approach emotion detection. Three MobileNetV2-RNN, InceptionV3-RNN, custom CNN-RNN for classification. achieved an accuracy rate 63%, while MobileNetV2-RNN InceptionV3-RNN transfer learning yield 59% 66%, respectively. demonstrate enhanced efficiency distinguishing these nuanced emotions, significant advancement field expression recognition. holds substantial implications cognitive real-world applications, particularly enhancing interactive digital emotional analysis.