Enhancing Emotion Detection from Facial Expressions using Advanced CNN Architectures

Abstract

Emotion detection plays a vital role in human–computer interaction, affective computing, and intelligent surveillance systems. Relying on traditional methods or handcrafted features often leads to limited performance due to variations in lighting, pose, and facial expressions. To address these challenges, this work proposes a CNN-based facial emotion detection framework that focuses solely on facial expressions for accurate emotion recognition. The proposed system utilizes Convolutional Neural Networks (CNNs) to automatically extract high-level discriminative features from facial images. Input images are preprocessed through resizing and normalization to ensure consistency, and data augmentation techniques are applied to improve generalization. The CNN model learns spatial features such as edges, textures, and facial muscle movements through multiple convolutional and pooling layers, eliminating the need for manual feature extraction. The extracted features are passed through fully connected layers followed by a Softmax classifier to predict emotional states such as happy, sad, angry, fear, disgust, surprise, and neutral. This approach effectively captures complex facial patterns and enhances classification performance. Experimental evaluation shows that the CNN-based model achieves higher accuracy, better generalization, and improved robustness compared to traditional machine learning methods. The results demonstrate that the proposed system is well-suited for real-world applications such as human–computer interaction, surveillance systems, healthcare monitoring, and affect-aware intelligent systems