Applying artificial intelligence models for the automatic forest fire detection
Autor: | El Madafri, Ismail, Peña Carrera, Marta, Olmedo Torre, Noelia |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2023 |
Předmět: |
Desenvolupament humà i sostenible::Medi ambient::Ecologia [Àrees temàtiques de la UPC]
Graphic-based intelligent surveillance Forest fires Deep learning (Machine learning) Incendis forestals Informàtica::Intel·ligència artificial [Àrees temàtiques de la UPC] Deep learning Models optimization Forest fire detection Sustainable management Sistemes de gestió mediambiental Transfer learning Aprenentatge profund |
Popis: | Throughout the past decade, the development of Artificial Intelligence-based devices for the automatic detection of early stage forest fires has been a growing focus. Computer Vision techniques are well-suited for this problem due to the distinctive visual characteristics of forest fires. The effectiveness of several Artificial Intelligence algorithms in a binary classification problem involving fire/ non-fire images was assessed by comparing them using a publicly available dataset. The benchmark dataset was used to both train and evaluate the models. An optimization method was employed to train the Artificial Intelligence algorithms, resulting in a higher performance than that previously achieved by studies on the same dataset. |
Databáze: | OpenAIRE |
Externí odkaz: |