SagarLekhak/pfm1-landmine-uav-vnir-hsi-IGARSS-2026

---
license: cc-by-4.0
task_categories:
- object-detection
- image-segmentation
- image-classification
tags:
- remote-sensing
- hyperspectral
- remote_sensing
- IGARSS2026
- VNIR
---

# <p align="center"> PFM-1 Landmine VNIR Hyperspectral Imaging Dataset (IGARSS 2026)</p>

<p align="center">
<a href="https://arxiv.org/abs/2602.10434"><img src="https://img.shields.io/badge/Paper-IGARSS%202026-blue" alt="Paper"></a>
<a href="https://github.com/PrasannaPulakurthi/pfm1-hsi-benchmark"><img src="https://img.shields.io/badge/GitHub-pfm1--hsi--benchmark-green" alt="Dataset"></a>
<a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-yellow.svg" alt="License: MIT"></a>
</p>

## Dataset Description

This dataset [1](#ref1) contains Visible and Near-Infrared (VNIR) Hyperspectral Imaging (HSI) data prepared for the
IEEE International Geoscience and Remote Sensing Symposium (IGARSS) 2026.

This specific version is a refined subset of the original benchmark dataset [2](#ref2). While the original
release provided full radiance cubes, broad GCP/AeroPoint data, and reference ground spectra of all the targets,
this version focuses on a spatially subsetted region containing only PFM-1 landmine targets and
includes high-accuracy, pixel-wise binary ground truth masks.

The data was collected over a controlled test field seeded with 143 realistic surrogate landmine and
UXO targets (surface, partially buried, and fully buried). Data acquisition was performed using
a Headwall Nano-Hyperspec® sensor mounted on a multi-sensor UAV platform flown at an altitude of ~20.6 m [2](#ref2).

For more details regarding data acquistion and preprocessing, go to the original paper [2](#ref2).

- **Sensor:** [Headwall Nano-Hyperspec®]
- **Spectral Range:** [398–1002 nm]
- **Number of Bands:** [270 bands]
- **Approximate GSD:** [Approx. 1.29 cm]

## Dataset File Structure

The dataset is organized as follows:

| File Name | Size | Description |
| :--- | :--- | :--- |
| `site_with_only_mines` | 6.42 GB | Main Hyperspectral (HSI) data cube (ENVI format), referred to as "Full Region" in the paper [1]. |
| `site_with_only_mines.hdr` | 23.2 kB | Header file containing metadata for the HSI data cube. |
| `roi_site_with_only_mines.roi` | 256 B | Region of Interest (ROI) file used for spatial subsetting the original dataset from [2]. |
| `binary_ground_truth_mask_for_pfm1` | 5.9 MB | Pixel-wise binary ground truth mask for PFM-1 mine locations. |
| `binary_ground_truth_mask_for_pfm1.hdr` | 803 B | Header file for the binary ground truth mask. |
| `target_signature_pfm_1_svc.txt` | 7.86 kB | Reference ground spectral signature for PFM-1 (captured via SVC). |
| `.gitattributes` | 2.63 kB | Configuration file for Git LFS and XetData tracking. |

## Usage

### Download
```bash
pip install huggingface_hub
```
```python
from huggingface_hub import snapshot_download

snapshot_download(repo_id="SagarLekhak/pfm1-landmine-uav-vnir-hsi-IGARSS-2026", repo_type="dataset", local_dir="./data")
```

### To load the data using Python
```python
import spectral.io.envi as envi

# Load the hyperspectral cube
img = envi.open('site_with_only_mines.hdr', 'site_with_only_mines')
print(f"Loaded cube with {img.shape[2]} spectral bands.")
```

## Citation
If you use this dataset, please cite the specific IGARSS 2026 work [1](#ref1) and the original benchmark paper [2](#ref2):

<a name="ref1"></a>
**[1] IGARSS 2026 Work:**
```bibtex
@misc{lekhak2026benchmarkingdeeplearningstatistical,
title={Benchmarking Deep Learning and Statistical Target Detection Methods for PFM-1 Landmine Detection in UAV Hyperspectral Imagery},
author={Sagar Lekhak and Prasanna Reddy Pulakurthi and Ramesh Bhatta and Emmett J. Ientilucci},
year={2026},
eprint={2602.10434},
archivePrefix={arXiv},
primaryClass={eess.IV},
url={https://arxiv.org/abs/2602.10434},
}
```

<a name="ref2"></a>
**[2] Original Benchmark Dataset:**
```bibtex
@misc{lekhak2026uavbasedvnirhyperspectralbenchmark,
title={A UAV-Based VNIR Hyperspectral Benchmark Dataset for Landmine and UXO Detection},
author={Sagar Lekhak and Emmett J. Ientilucci and Jasper Baur and Susmita Ghosh},
year={2026},
eprint={2510.02700},
archivePrefix={arXiv},
primaryClass={eess.IV},
url={https://arxiv.org/abs/2510.02700},
}
```