Contactless Mask Monitoring and Biometric Attendance System

ABSTRACT

This paper attempts to give an overview of the system which is designed keeping social distancing guidelines in mind. Our system will detect in real-time, if the person in the captured live video is wearing a mask properly or not using a mask detecting algorithm developed using deep learning and neural networks with an accuracy of 96.05%. If and only if the person is wearing a mask, they will be allowed to scan the iris and hence record their attendance, which can be stored in excel or CSV formats. The location of iris biometric is translated to a real-life position in the 3D space with the resolution of 0.lmm. To scan the located biometric this system comprises a robotic arm. End effector of this robotic arm traverses to the translated position of the person's eye to scan iris with an iris scanner. The system employs a 'four degrees of motion' robotic arm that can autonomously align itself to the iris with an accuracy of 96.86%. It is battery operated and has a cylindrical workspace with maximum range of 300mm, hence it is easily deployable in institutions requiring secure authorization while monitoring COVID-19 safety norms. © 2022 IEEE.