TY - GEN
T1 - Smoke detection in video using a raspberry pi
AU - Santamaria-Guerrero, Diego
AU - Mejia-Flores, Karina
AU - Castro-Madrid, Luis Carlos
AU - Toscano-Medina, Rocio
AU - Prudente-Tixteco, Lidia
AU - Olivares-Mercado, Jesus
AU - Toscano-Medina, Karina
AU - Sanchez-Perez, Gabriel
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/10
Y1 - 2019/10
N2 - Many different algorithms have been developed for smoke detection of in video. However, most of these algorithms have been implemented in simulations with specialized software or computers with high computational capacity. This article presents results of implementation of a method to smoke detection in video, which was implemented in a Raspberry Pi embedded computer using OpenCV library. This method proposes to use different smoke characteristics such as: movement, color, direction, growth, accumulation and a background subtraction method, so that movement detection is as optimal as possible to be able to detect early smoke presence and prevent a possible mishap. This algorithm was evaluated using 22 different videos obtaining up to 90.9% accuracy in smoke detection and up to 9.1% false positives, showing that proposed method obtains favorable results.
AB - Many different algorithms have been developed for smoke detection of in video. However, most of these algorithms have been implemented in simulations with specialized software or computers with high computational capacity. This article presents results of implementation of a method to smoke detection in video, which was implemented in a Raspberry Pi embedded computer using OpenCV library. This method proposes to use different smoke characteristics such as: movement, color, direction, growth, accumulation and a background subtraction method, so that movement detection is as optimal as possible to be able to detect early smoke presence and prevent a possible mishap. This algorithm was evaluated using 22 different videos obtaining up to 90.9% accuracy in smoke detection and up to 9.1% false positives, showing that proposed method obtains favorable results.
KW - Embedded systems
KW - Image processing
KW - Move detection
KW - Smoke detection
KW - Video processing
UR - http://www.scopus.com/inward/record.url?scp=85077188117&partnerID=8YFLogxK
U2 - 10.1109/ICEV.2019.8920604
DO - 10.1109/ICEV.2019.8920604
M3 - Contribución a la conferencia
AN - SCOPUS:85077188117
T3 - 2019 IEEE International Conference on Engineering Veracruz, ICEV 2019
BT - 2019 IEEE International Conference on Engineering Veracruz, ICEV 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE International Conference on Engineering Veracruz, ICEV 2019
Y2 - 14 October 2019 through 17 October 2019
ER -