Deep-learning-based grassland mapping with Sentinel-2: prioritizing key spectral bands and time periods
Journal
SPIE
Date Issued
September 19, 2025
Abstract
Accurate grassland mapping is essential for biodiversity conservation and sustainable land management, yet
remains challenging due to the spectral and temporal variability of grassland ecosystems. This study presents
a Deep Learning approach for grassland classification using multi-temporal Sentinel-2 imagery, incorporating
a dynamic feature selection mechanism to prioritize informative spectral bands and time periods. In order to
allow the model to adaptively focus on discriminative temporal spectral patterns, we compare a baseline neural
network with a modified design that learns to weight input features dynamically. Our findings demonstrate
that the feature selection model achieves superior performance (Accuracy: 0.954 ± 0.004, MCC: 0.726 ± 0.027)
compared to both the baseline network and single-date models, highlighting the importance of temporal diversity in grassland classification.
remains challenging due to the spectral and temporal variability of grassland ecosystems. This study presents
a Deep Learning approach for grassland classification using multi-temporal Sentinel-2 imagery, incorporating
a dynamic feature selection mechanism to prioritize informative spectral bands and time periods. In order to
allow the model to adaptively focus on discriminative temporal spectral patterns, we compare a baseline neural
network with a modified design that learns to weight input features dynamically. Our findings demonstrate
that the feature selection model achieves superior performance (Accuracy: 0.954 ± 0.004, MCC: 0.726 ± 0.027)
compared to both the baseline network and single-date models, highlighting the importance of temporal diversity in grassland classification.
File(s)![Thumbnail Image]()
Name
1381610.pdf
Size
390.5 KB
Format
Adobe PDF
Checksum (MD5)
99a4b80431c932028b13a24ec839cf29