FACE EXPRESSION RECOGNIZER DENGAN CONVOLUTIONAL NEURAL NETWORK UNTUK MEMBANTU PENDERITA AUTISME MENGENALI EKSPRESI WAJAH SESEORANG

Abstract

Autism is a type of mental health disorder that is 80% caused by heredity, and environmental influences cause the rest. People with autism tend to be unable to concentrate and look at the other person when interacting. Since the age of toddlers, people with autism have a shallow response to the surrounding environment. In addition, people with autism also find it very difficult to recognize someone's expression, even though facial expressions or facial expressions are one way that can be used to recognize someone's emotions. In addition, facial expressions indirectly reveal the contents of a person's thoughts. To overcome this, the author wants to build a face expression recognition model to help people with autism recognize someone's facial expressions. The primary purpose of this research is to help people with autism socialize and recognize the facial expressions of the people around them. This face expression recognition model was built by applying Convolutional Neural Network (CNN) intelligence and using the Tensorflow library and the Keras API. The dataset used in this study is a collection of faces from all over the world. In this research and model development process, the output display of the detection of facial expressions is in the form of diagrams and descriptions of the expressions that a person is experiencing.