Uncontrolled wildfires are dangerous events capable of harming people safety. To contrast their increasing impact in recent years, a key task is an accurate detection of the affected areas and their damage assessment from satellite images. Current state-of-the-art solutions address such problem through a double convolutional neural network able to automatically detect wildfires in satellite acquisitions and associate a damage index from a defined scale. However, such deep-learning model performance is strongly dependent on many factors. In this work, we specifically focus on a key parameter, i.e., the loss function, exploited in the underlying neural networks. Besides the state-of-the-art solutions based on the Dice-MSE, among the many loss functions proposed in literature, we focus on the Binary Cross-Entropy (BCE) and the Intersection over Union (IoU), as two representatives of the distribution-based and region-based categories, respectively. Experiments show that the BCE loss function coupled with a double-step U-Net architecture provides better results than current state-of-the-art solutions on a public labeled dataset of European wildfires.

Improving Wildfire Severity Classification of Deep Learning U-Nets from Satellite Images / Monaco, Simone; Pasini, Andrea; Apiletti, Daniele; Colomba, Luca; Garza, Paolo; Baralis, ELENA MARIA. - ELETTRONICO. - (2020), pp. 5786-5788. (Intervento presentato al convegno 2020 IEEE International Conference on Big Data tenutosi a Atlanta (US) nel December 10-13, 2020) [10.1109/BigData50022.2020.9377867].

Improving Wildfire Severity Classification of Deep Learning U-Nets from Satellite Images

Simone Monaco;Andrea Pasini;Daniele Apiletti;Luca Colomba;Paolo Garza;Elena Baralis
2020

Abstract

Uncontrolled wildfires are dangerous events capable of harming people safety. To contrast their increasing impact in recent years, a key task is an accurate detection of the affected areas and their damage assessment from satellite images. Current state-of-the-art solutions address such problem through a double convolutional neural network able to automatically detect wildfires in satellite acquisitions and associate a damage index from a defined scale. However, such deep-learning model performance is strongly dependent on many factors. In this work, we specifically focus on a key parameter, i.e., the loss function, exploited in the underlying neural networks. Besides the state-of-the-art solutions based on the Dice-MSE, among the many loss functions proposed in literature, we focus on the Binary Cross-Entropy (BCE) and the Intersection over Union (IoU), as two representatives of the distribution-based and region-based categories, respectively. Experiments show that the BCE loss function coupled with a double-step U-Net architecture provides better results than current state-of-the-art solutions on a public labeled dataset of European wildfires.
2020
978-1-7281-6251-5
File in questo prodotto:
File Dimensione Formato  
IEEE_Conference_BigData_2020.pdf

accesso aperto

Descrizione: Accepted article
Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 117.32 kB
Formato Adobe PDF
117.32 kB Adobe PDF Visualizza/Apri
09247550.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 6.07 MB
Formato Adobe PDF
6.07 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2853627