AMLID: An Adaptive Multispectral Landmine Identification Dataset for Drone-Based Detection

2025-12-21

AI Summary

Existing mine detection methods suffer from high operational risk, low efficiency, and prohibitive costs. To address these challenges, this paper introduces the first open-source UAV-based multispectral mine recognition benchmark dataset. It integrates synchronized RGB and long-wave infrared (LWIR) imagery, encompassing 21 real-world mine types, 11 fusion modalities, four flight altitudes, and diverse seasonal and illumination conditions—totaling 12,078 precisely annotated images. Our key contributions are threefold: (1) the first publicly available, fully controlled, dual-modal (RGB-LWIR) dataset with systematic variation across altitude, season, illumination, casing material, and mine type (anti-personnel/anti-tank); (2) enabling safe, hardware-agnostic algorithm development without live ordnance or specialized equipment; and (3) facilitating rigorous cross-scenario generalization evaluation. This benchmark significantly lowers entry barriers for research and advances the robustness and transferability of UAV-based mine detection algorithms in complex, real-world field environments.

Technology Category

• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Landmines remain a persistent humanitarian threat, with an estimated 110 million mines deployed across 60 countries, claiming approximately 26,000 casualties annually. Current detection methods are hazardous, inefficient, and prohibitively expensive. We present the Adaptive Multispectral Landmine Identification Dataset (AMLID), the first open-source dataset combining Red-Green-Blue (RGB) and Long-Wave Infrared (LWIR) imagery for Unmanned Aerial Systems (UAS)-based landmine detection. AMLID comprises of 12,078 labeled images featuring 21 globally deployed landmine types across anti-personnel and anti-tank categories in both metal and plastic compositions. The dataset spans 11 RGB-LWIR fusion levels, four sensor altitudes, two seasonal periods, and three daily illumination conditions. By providing comprehensive multispectral coverage across diverse environmental variables, AMLID enables researchers to develop and benchmark adaptive detection algorithms without requiring access to live ordnance or expensive data collection infrastructure, thereby democratizing humanitarian demining research.

Problem

• Develops a dataset for drone-based landmine detection using multispectral imagery
• Addresses hazardous and costly current landmine detection methods
• Enables algorithm development without live ordnance or expensive infrastructure

Innovation

• First open-source RGB-LWIR dataset for drone detection
• Includes 12,078 labeled images across 21 landmine types
• Covers multiple fusion levels, altitudes, seasons, and lighting