Design and Architecture of a Public Satisfaction Detection Camera Based on Facial Emotional Analysis

Abstract

One way to measure customer satisfaction is to involve the Internet of Things and Artificial Intelligence in the form of mood recognition services provided by Amazon Web Service and the services available to it, such as S3 (Simple Storage Service, a storage service from AWS), Relational Database Services, Rekognition services (An Image Recognition Services), Application Protocol Interface, Lambda Services. The source of input acquisition in this study uses the ESP-EYE device, the ESP32 microcontroller which has a camera. The device is connected to Amazon web service via the internet. Data is stored in a storage service called S3. S3 is used as a storage bucket, where Lambda will be the center of the process that pre-processes the face image data to crop the edge. After the cropping process, lambda will call the Rekognition service to start the Rekognition services process. The relationship of each data is stored in RDS using MariaDB. The results are, that the internet of Customer Satisfaction Systems can be built with the involvement by several products of Amazon Web Services. There is also considerable efficiency after the face image file has been processed by lambda to crop. The output from the system is recorded in the database and sent back to ESP-EYE for display in the serial monitor. The development of this system can be forwarded to an information system that allows each column in the database to be presented in the form of an information system. This system can help to measure customer satisfaction and recapitulate the results without involving the surveyor team so that it will save operational costs and time efficiency.