DESIGN AND IMPLEMENTATION OF AN IMPROVED MODEL FOR FACIAL, EXPRESSION DETECTION SYSTEM USING CNN

In recent years, the detection of facial emotions has gained considerable attention as it has diverse applications in fields like psychology, human-computer interaction, and marketing. To enhance the accuracy of facial emotion detection systems, convolutional neural networks (CNNs) have shown immense potential. This study presents a CNN-based approach for facial emotion detection that uses transfer learning and data augmentation techniques to boost the generalization of the model. The FER-2013 dataset was used to complete the facial expression recognition, the proposed technique can achieve an expression recognition rate of up to 99.68 percent aimed at the dataset being used. The results indicate that the CNNbased approach is highly effective in accurately recognizing facial emotions and has significant potential for real-world applications of facial emotion detection. Keywords: facial emotion, convolutional neural network, FER-2013.