Anxiety Recognition from Facial Expressions using Key Point Algorithm and Linear Support Vector Machine

Recognizing human facial expressions and emotions is a fascinating and difficult topic. Facial expressions play a significant part in emotion perception and are employed in nonverbal communication between people. Facial expressions are crucial in everyday emotional communication since they are an indicator of sentiments and allow a person to convey his emotional state. People can instantly understand another person's emotional condition based on their facial expressions. As a result, information on facial expressions is frequently employed in automatic emotion recognition systems. Anxiety is one of the most prevalent emotions that people experience in various situations. As a result, it must be identified and treated. The goal of this work is to demonstrate an automated anxiety recognition system based on facial expressions employing key point techniques. SIFT, ORB, and SURF features are used to construct facial feature vectors from key point descriptors. The results revealed that using the Key Point feature significantly enhances the process of detecting face characteristics. The results indicate the effectiveness of our proposed algorithms, notably Shift technology. In summary, the Shift algorithm has the highest discrimination rate, although it takes longer to execute than other technologies.