Emotion Detection System for Children with Autism Spectrum Disorder based on IoT and Deep learning

Today, parents of children are feeling difficult having problem of autism spectrum disorder since their child struggle to recognize and express their emotions, which makes them feel alone. The design and execution of an emotion detection system for kids with ASD are presented in this research. The system uses python libraries such as open cv and haar cascade method to detect and predict facial expressions in real time. Unlike most existing systems that focus on social media like Snapchat, this project is centered on real-time facial expression recognition (FER) aimed at improving the social interaction of those suffering from such disorders such as autism spectrum disorder (ASD). The child's face is captured by the system in real time so as to capture her emotional state from her facial gestures. It also responds to detected emotions like playing a song if it notices a happy face to improve the social behaviour of kids with ASD. For example, an engaging audio track may be played when it identifies a smiley face. This helps maintain calmness in children thus fostering interactive learning through an ATmega microcontroller coupled with DF mini-player. A stress sensor based on Galvanic Skin Response (GSR) has also been added for monitoring purposes of stress levels within this system. For continuous monitoring and management purpose stress data is transmitted through Node MCU (ESP8266-12E microcontroller) over Internet in real time.