Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area Delineation
In crisis management and remote sensing, image segmentation plays a crucial role, enabling tasks like disaster response and emergency planning by analyzing visual data. Neural networks are able to analyze satellite acquisitions and determine which areas were affected by a catastrophic event. The pro...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10980409/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849732124383379456 |
|---|---|
| author | Daniele Rege Cambrin Luca Colomba Paolo Garza |
| author_facet | Daniele Rege Cambrin Luca Colomba Paolo Garza |
| author_sort | Daniele Rege Cambrin |
| collection | DOAJ |
| description | In crisis management and remote sensing, image segmentation plays a crucial role, enabling tasks like disaster response and emergency planning by analyzing visual data. Neural networks are able to analyze satellite acquisitions and determine which areas were affected by a catastrophic event. The problem in their development in this context is the data scarcity and the lack of extensive benchmark datasets, limiting the capabilities of training large neural network models. In this article, we propose a novel methodology, namely Magnifier, to improve segmentation performance with limited data availability. The Magnifier methodology is applicable to any existing encoder–decoder architecture, as it extends a model by merging information at different contextual levels through a dual-encoder approach: a local and global encoder. Magnifier analyzes the input data twice using the dual-encoder approach. In particular, the local and global encoders extract information from the same input at different granularities. This allows Magnifier to extract more information than the other approaches given the same set of input images. Magnifier improves the quality of the results of +2.65% on average intersection over union while leading to a restrained increase in terms of the number of trainable parameters compared to the original model. We evaluated our proposed approach with state-of-the-art burned area segmentation models, demonstrating, on average, comparable or better performances in less than half of the giga floating point operations per second (GFLOPs). |
| format | Article |
| id | doaj-art-08d42909012e42e3ac1c38cc126cc8f6 |
| institution | DOAJ |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-08d42909012e42e3ac1c38cc126cc8f62025-08-20T03:08:20ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118122631227710.1109/JSTARS.2025.356581910980409Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area DelineationDaniele Rege Cambrin0https://orcid.org/0000-0002-5067-2118Luca Colomba1https://orcid.org/0000-0003-2911-4522Paolo Garza2https://orcid.org/0000-0002-1263-7522Politecnico di Torino, Turin, ItalyPolitecnico di Torino, Turin, ItalyPolitecnico di Torino, Turin, ItalyIn crisis management and remote sensing, image segmentation plays a crucial role, enabling tasks like disaster response and emergency planning by analyzing visual data. Neural networks are able to analyze satellite acquisitions and determine which areas were affected by a catastrophic event. The problem in their development in this context is the data scarcity and the lack of extensive benchmark datasets, limiting the capabilities of training large neural network models. In this article, we propose a novel methodology, namely Magnifier, to improve segmentation performance with limited data availability. The Magnifier methodology is applicable to any existing encoder–decoder architecture, as it extends a model by merging information at different contextual levels through a dual-encoder approach: a local and global encoder. Magnifier analyzes the input data twice using the dual-encoder approach. In particular, the local and global encoders extract information from the same input at different granularities. This allows Magnifier to extract more information than the other approaches given the same set of input images. Magnifier improves the quality of the results of +2.65% on average intersection over union while leading to a restrained increase in terms of the number of trainable parameters compared to the original model. We evaluated our proposed approach with state-of-the-art burned area segmentation models, demonstrating, on average, comparable or better performances in less than half of the giga floating point operations per second (GFLOPs).https://ieeexplore.ieee.org/document/10980409/Deep learningearth observation (EO)natural hazard managementpostwildfire segmentationsemantic segmentation |
| spellingShingle | Daniele Rege Cambrin Luca Colomba Paolo Garza Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area Delineation IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning earth observation (EO) natural hazard management postwildfire segmentation semantic segmentation |
| title | Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area Delineation |
| title_full | Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area Delineation |
| title_fullStr | Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area Delineation |
| title_full_unstemmed | Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area Delineation |
| title_short | Magnifier: A Multigrained Neural Network-Based Architecture for Burned Area Delineation |
| title_sort | magnifier a multigrained neural network based architecture for burned area delineation |
| topic | Deep learning earth observation (EO) natural hazard management postwildfire segmentation semantic segmentation |
| url | https://ieeexplore.ieee.org/document/10980409/ |
| work_keys_str_mv | AT danieleregecambrin magnifieramultigrainedneuralnetworkbasedarchitectureforburnedareadelineation AT lucacolomba magnifieramultigrainedneuralnetworkbasedarchitectureforburnedareadelineation AT paologarza magnifieramultigrainedneuralnetworkbasedarchitectureforburnedareadelineation |