A Review on Fire Detection Architectures and Techniques in Indoor and Outdoor Spaces
Tipo:
Articulos de Divulgación
Autor:
Caballero-Hernández, H.*, Muñoz-Jiménez, V., Ramos-Corchado, M.A., Gil-Antonio, L.
Fecha:
2026-01-01
Descripción:
Fires in residential and forested areas can escalate rapidly, causing catastrophic consequences such as partial or total destruction, loss of life, economic damage, and air pollution. This review compiles research on fire detection in outdoor and indoor environments. The surveyed approaches leverage computer vision, machine learning (ML), and sensing systems that measure variables such as smoke, CO, and CO2, often validated through wireless sensor networks (WSN) and geographic information systems (GIS). Satellite-based monitoring is also discussed as a complementary solution that uses sensors and artificial intelligence (AI) algorithms. Across the literature, AI significantly improves the reliability of detection and prediction in diverse contexts; convolutional neural networks (CNNs) are among the most frequently used methods, and are commonly integrated with sensor networks, satellite platforms, and unmanned autonomous vehicles.
Keywords: Computer vision, Deep Learning, Heat point, Wildfire, Wireless Sensor Networks.
;)
