REAL TIME HUMAN FACE EXPRESSION RECOGNITION

Computer Engineering

This study proposes a system for facial expression recognition using Convolutional Neural Networks (CNNs) trained on the FER2013 dataset. The objective is to accurately classify human emotions from facial expressions, including anger, disgust, fear, happiness, sadness, surprise, and neutral. The CNN model is trained on FER2013, a dataset that contains over 35,000 labeled images of human faces displaying one of the seven emotions. The CNN model extracts features from the input image using convolutional and pooling layers, followed by fully connected layers for emotion prediction. The model is trained by adjusting the weights to minimize the difference between the predicted and true emotions. The system is evaluated using a separate set of data called the validation set. The validation accuracy of the proposed system was found to be 60%. Facial expression recognition using CNNs has numerous applications, such as in human computer interaction, virtual reality, and security systems. However, challenges remain in handling variations in lighting, pose, and occlusion, which can affect the accuracy of the system. The proposed system contributes to the field of computer vision by offering a viable approach for emotion recognition and has the potential to be further developed for real-world applications.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.