MineInsight: A Multi-Sensor Dataset for Humanitarian Demining Robotics in Off-Road Environments
Mario Malizia, Charles Hamesse, Ken Hasselmann, Geert De Cubber, Nikolaos Tsiogkas, Eric Demeester, Rob Haelterman
AI summary
Problem
Existing landmine detection datasets suffer from narrow sensor diversity, lack of standardized formats, and insufficient coverage of challenging off-road conditions, which hinders the reliable validation of robotic detection algorithms.
Approach
The team deployed a UGV with a synchronized robotic arm carrying LiDAR and multi-spectral cameras across three diverse outdoor tracks to collect, calibrate, and annotate a comprehensive dataset of landmines and common objects.
Key results
- First dataset integrating dual-view UGV and robotic arm sensor scans
- Collection of approximately 196,000 synchronized RGB, VIS-SWIR, and LWIR frames
- Inclusion of 15 inert landmine types and 20 common false-positive objects across three distinct terrains
- Provision of human-refined 2D bounding boxes and standardized ROS 2 formatted data
Why it matters
Enables researchers and humanitarian demining organizations to develop and validate robust, multi-sensor detection algorithms for safer and more efficient robotic land clearance.
Abstract
The use of robotics in humanitarian demining increasingly involves computer vision techniques to improve landmine detection capabilities. However, in the absence of diverse and realistic datasets, the reliable validation of algorithms remains a challenge for the research community. In this paper, we introduce MineInsight, a publicly available multi-sensor, multi-spectral dataset designed for off-road landmine detection. The dataset features 35 different targets (15 landmines and 20 commonly found objects) distributed along three distinct tracks, providing a diverse and realistic testing environment. MineInsight is, to the best of our knowledge, the first dataset to integrate dual-view sensor scans from both an Unmanned Ground Vehicle and its robotic arm, offering multiple viewpoints to mitigate occlusions and improve spatial awareness. It features two LiDARs, as well as images captured at diverse spectral ranges, including visible (RGB, monochrome), visible short- wave infrared (VIS-SWIR), and long-wave infrared (LWIR). Additionally, the dataset provides bounding boxes generated by an automated pipeline and refined with human supervision. We recorded approximately one hour of data in both daylight and nighttime conditions, resulting in around 38,000 RGB frames, 53,000 VIS-SWIR frames, and 108,000 LWIR frames. MineInsight serves as a benchmark for developing and evalu- ating landmine detection algorithms. Our dataset is available at https://github.com/mariomlz99/MineInsight.