SAE-LWIR is the first publicly available dataset generated with MODTRAN for atmospheric compensation in standoff long-wave infrared hyperspectral imaging. The dataset supports the paper 'Set-Based Transformer for Atmospheric Compensation in Standoff LWIR Hyperspectral Imaging' presented at IGARSS 2026. It was created by researchers from the Universidad Industrial de Santander in Bucaramanga, Colombia.
Use Cases
- Training atmospheric compensation models based on MODTRAN-simulated long-wave infrared hyperspectral data.
- Benchmarking new algorithms for standoff hyperspectral image analysis based on synthetic atmospheric effects.
- Developing transformer-based architectures for remote sensing tasks based on the dataset's described application.
Strengths
- Dataset is described as the first publicly available of its kind generated with MODTRAN.
- Associated with peer-reviewed research presented at the IGARSS 2026 conference.
Limitations
- Description metadata is limited; actual data quality requires manual inspection after download.
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
Provenance
- Source
- SemilleroCV
- Collection Method
- Generated using the MODTRAN atmospheric radiative transfer code.
- Freshness
- Last updated 2026-06-06 19:59:31; freshness should be verified.