Facial Expression Classification Analysis Using Facial Images Based on Resnet-50V2

Emotion is one of the ways to express feelings, it can be shown through facial expressions or body movements. One of the easiest ways to tell someone's emotions is by looking at their facial expressions. But to detect emotions on human faces using computers and get high accuracy results is a challenge for computers to do. Facial emotional expression works with color segments and facial points that form patterns. Facial emotions work to describe the condition of the sender of the message to the recipient of the message. Emotions have a role as a supporting indicator in communicating in addition to the intonation of speech. The purpose of this research is to classify human emotions using Convolutional Neural Network (CNN) with ResNet-50v2 architecture. The best result is 83.16% using 200 epochs.