Facial Emotion Recognition (FER) has become one of the most active research areas in the fields of Artificial Intelligence and Computer Vision due to its wide range of applications in healthcare, education, human-computer interaction, security, and intelligent monitoring systems.The ability of computers to automatically interpret human emotions from facial expressions has significantly improved with the rapid development of machine learning and deep learning techniques.This review paper presents a comprehensive analysis of recent advances in facial emotion recognition systems, with particular attention to deep learning-based approaches.The study discusses the main stages of FER systems, including image acquisition, preprocessing, feature extraction, and classification.In addition, widely used public datasets such as FER2013, CK+, RAF-DB, and AffectNet are reviewed and compared based on their characteristics and research challenges.The paper further examines the transition from traditional machine learning algorithms to advanced convolutional neural network architectures that have achieved higher recognition accuracy in complex real-world scenarios.Moreover, critical challenges such as class imbalance, illumination variations, occlusions, head pose changes, and the recognition of subtle micro-expressions are analyzed.Finally, the paper highlights emerging research directions, including explainable artificial intelligence, multimodal emotion recognition, lightweight real-time systems, and transformer-based architectures.This review aims to provide researchers and practitioners with a clear overview of current developments and future opportunities in facial emotion recognition research.
Zeqo et al. (Thu,) studied this question.