Efficient Mask Detection System for Low Powered Micro-Computer
5th IEEE International Conference on Computational Systems and Information Technology for Sustainable Solutions, CSITSS 2021
; 2021.
Article
in English
| Scopus | ID: covidwho-1741148
ABSTRACT
Wearing masks has been one of the key methods of preventing the spread of COVID-19. Being able to ensure that the people entering the premises of any institution are wearing masks reduces the risk of the people within those premises to be affected by the virus. This paper describes the approach of taking Google's pre-trained Inception-V3 architecture and using transfer learning to adapt it for mask detection in Indian scenario. The mask detection model was converted to a TFlite version deployable on Raspberry Pi 3 Model B. Pi camera was used for data capture. All hardware chosen and software adaptations were done with the focus of making the model portable and affordable. This model was created with the focus of providing a cost-efficient way to enforce preventive measures during this pandemic. © 2021 IEEE.
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Language:
English
Journal:
5th IEEE International Conference on Computational Systems and Information Technology for Sustainable Solutions, CSITSS 2021
Year:
2021
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS