Autism Spectrum Disorder Detection Using Transfer Learning with VGG 19, Inception V3 and DenseNet 201

A person with autism or autism spectrum disorder (ASD) has trouble recognizing, socializing, and communicating with others. ASD is not only about one situation. In fact, it has numerous groups of situations. Although it cannot be fully recovered, the right treatments and services can help a person's symptoms and daily activities. Deep learning has achieved outstanding results in pattern recognition tasks in the current times. CNN-based methods are widely suggested for the research. By taking proper care, children can improve easily. For this, we decided to work on autism in children through image classification for early detection. Early detection can help children provide strength and capability for a better life. It can lessen children's symptoms and can improve their normal development by assisting them to analyze new capabilities as a way to allow them to be independent for the rest of their lives. To deal with these difficulties, the right guidelines and strategies can help a lot. To provide the necessary guidelines in time, early detection is very important. We employed deep learning techniques to conduct our research utilizing the image dataset. We tried to implement a method that can detect autism in children through facial expressions. In this research, we use VGG 19, Inception V3, and DenseNet 201.