ADVANCED HUMAN EMOTION DETECTION USING FACIAL FEATURES AND DIGITAL IMAGE PROCESSING TECHNIQUES

Abstract

This research presents a novel Human Emotion Detection utilizing Facial Features (HEDFF) framework that integrates deep learning, optimization techniques, and advanced image processing for highly accurate emotion recognition. The proposed framework employs the AffectNet Dataset, a large-scale and diverse dataset containing richly annotated facial expressions, to enhance the robustness and generalization capability of the model. A key novelty of this study lies in the hybrid integration of Gabor filtering, Recurrent Neural Networks (RNNs), EfficientNet architecture, and Particle Swarm Optimization (PSO) within a unified emotion detection pipeline. During preprocessing, Gabor filtering effectively extracts discriminative texture and edge information from facial images, improving feature representation under varying illumination and expression conditions. The RNN component captures temporal dependencies and sequential facial dynamics, enabling better understanding of subtle emotional transitions. Furthermore, the EfficientNet model provides a computationally efficient yet highly accurate classification mechanism through optimized scaling of network depth, width, and resolution. To further enhance system performance, PSO is applied for automated hyperparameter optimization, reducing computational complexity while maximizing classification accuracy. Unlike conventional emotion recognition systems that rely solely on static facial features or single deep learning architectures, the proposed HEDFF framework combines spatial, temporal, and optimization-based learning strategies to achieve superior robustness and precision. Experimental evaluations conducted on benchmark datasets demonstrate the effectiveness of the proposed approach, achieving a remarkable accuracy of 98.87%, precision of 97.11%, sensitivity of 95.15%, and specificity of 96.96%, outperforming several existing state-of-the-art methods. The proposed framework offers significant contributions to the fields of affective computing, intelligent human-computer interaction, healthcare monitoring, mental health assessment, and real-time emotion-aware applications.