Abstract
Emotion care for human well-being is important for all ages. In this paper, we propose an emotion care system based on big data analysis for autism disorder patient training, where emotion is detected in terms of facial expression. The expression can be captured through a camera as well as Internet of Things (IoT)-enabled devices. The system works with deep learning techniques on emotional big data to extract emotional features and recognize six kinds of facial expressions in real-time and offline. A convolutional neural network (CNN) model based on MobileNet V1 structure is trained with two emotional datasets, FER-2013 dataset and a new proposed dataset named MCFER. The experiments on three strategies showed that the proposed system with deep learning model obtained an accuracy of 95.89%. The system can also detect and track multiple faces as well as recognize facial expressions with high performance on mobile devices with a speed of up to 12 frames per second.
Highlights
With the proliferation of Artificial Intelligence (AI) and Internet of Things (IoT)-oriented medical applications, emotion plays an important role in human life and communication [1], [2]
To train a model for facial expression recognition (FER), all images of the datasets have to be detected by using the haar cascade classifier in order to determine if there is a face in the image
Based on the first stage pre-trained model on FER2013 dataset, the accuracy of model on MCFER dataset in the second stage is about 95.89%
Summary
With the proliferation of Artificial Intelligence (AI) and Internet of Things (IoT)-oriented medical applications, emotion plays an important role in human life and communication [1], [2]. Emotion is an inextricable part of the interaction of human beings, which can be observed by the changes in physiological features and behaviors. With the development of big data and deep learning, huge amount of data including emotional data is generated in recent years, which can not be handled with the traditional techniques. To this end, deep learning techniques has the potential to solve this problem [4]. By using deep learning techniques to analyze the emotional big data, machines can learn and understand emotions to meet human needs, because deep learning techniques can learn and track different physiological features on the body
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
