Summary - Y Suppl. 105 (05/2026) - ITU-T Y.4229 – Supplement on use of artificial intelligence models and datasets for smart forest firefighting systems

Recent observations show that the natural fire regime of several ecosystems has been modified by human activities and climate extremes. The significant advancements in the fields of image processing and machine learning (Recommendation ITU-T Y.3172) lead to the implementation of more accurate wildfire detection systems. However, several challenges remain to be addressed for effective fire detection, mainly for people living in urban or rural areas that are adjacent to these biomes. The performance of a wildfire detection system based on the requirements and reference functional model of Internet of things (IoT)-based smart forest firefighting system of Recommendation ITU-T Y.4229, also relies on the deep learning algorithms and the quality of the training dataset and its capacity to represent real-world scenarios.
This Supplement outlines challenges and opportunities associated with the use of smart forest firefighting systems and identifies and describes features of AI models and datasets for wildfire detection. In addition, this Supplement presents the application of these concepts through a wildfire detection use case in the Cerrado Savanna biome.