A low-cost embedded facial recognition system for door access control using deep learning

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

1 Cita (Scopus)

Resumen

This paper presents how the Neural Compute Stick 2 processor is used in conjunction with a Raspberry Pi 3 B+ board for controlling an electromagnetic lock using deep learning techniques to generate an access control system with embedded facial recognition capabilities. The system is based on object detection and facial recognition code developed in Python programming language. Through a webcam, it obtains images in real-time to make a comparison with the faces stored in a database. For prototype implementation, an embedded platform (Raspberry Pi) was used to provide the base operating system for programming and provide an analog signal in response to the system. A proof of concept experimentation was carried out with 15 subjects to test the effectiveness of the proposed method, achieving an average accuracy of 88.75%.

Idioma originalInglés
Título de la publicación alojada2020 IEEE ANDESCON, ANDESCON 2020
EditorialInstitute of Electrical and Electronics Engineers Inc.
ISBN (versión digital)9781728193656
DOI
EstadoPublicada - 13 oct. 2020
Evento2020 IEEE ANDESCON, ANDESCON 2020 - Quito, Ecuador
Duración: 13 oct. 202016 oct. 2020

Serie de la publicación

Nombre2020 IEEE ANDESCON, ANDESCON 2020

Conferencia

Conferencia2020 IEEE ANDESCON, ANDESCON 2020
País/TerritorioEcuador
CiudadQuito
Período13/10/2016/10/20

Huella

Profundice en los temas de investigación de 'A low-cost embedded facial recognition system for door access control using deep learning'. En conjunto forman una huella única.

Citar esto