Live Streaming of Face Mask and Body Temperature Detection System using Transfer Learning and IoT

Abstract

Since the outbreak of COVID-19 pandemic, enormous development has been done in computer vision systems in face masks and temperature detection. To enforce the mandate for wearing the mask in public places, the authors have suggested a solution using transfer learning (MobileNetV2) architecture as a foundation for image classification. The proposed solution is embedded devices with Raspberry Pi4 and python packages viz TensorFlow, OpenCV, and Keras. In this Work, a combination of transfer learning and IoT has been worked out using live streaming to detect an appropriate use of face masks along with body temperature. This technique includes the Single Shot Single box and Multi-box identification of the face mask. Further, a person can check the Temperature at the entrance. This temperature sensor is in noncontact mode, which senses a temperature from 2 cm to 5 cm and is displayed through LCD. It sends the data to the cloud for visualization and analysis, and an alert message is sent in a real-time environment based on that analysis.