Recognizing the emotion in the percent with the graylevel cosystality matrix and momentum propertiesbased on the K-NN classification method

In this thesis, we present an approach to recognizing emotion in a face, based on the idea of ensemble methods to classify seven different emotional states. In addition to the probabilistic fusion algorithm, action units and key point feature positions allow us to recognize seven basic emotions through facial expressions. Each sample is labeled as neutral, joy, sadness, anger, surprise, fear, or disgust. Separate neural network classifiers that extract two types of face features, action units, and feature point locations are trained together with a scaled combined backpropagation algorithm. Decision level fusion was performed to improve the performance of our system. K-NN was used for classification. KEYWORDS: Facial Emotion Detection, K-NN, Feature Extraction July 2021, 41 Page,