A Comprehensive Dataset for Underground Miner Detection in Diverse Scenario

2025-06-26

AI Summary

Underground mining environments present complex, light-deprived conditions, leading to severe data scarcity and poor algorithmic adaptability for automated miner detection. To address this, we introduce UMTID—the first benchmark thermal-infrared dataset specifically designed for miner detection in underground mines, covering diverse operational states (e.g., stationary, walking, crawling) and challenging environmental conditions (e.g., low illumination, dust, high temperature), thereby filling a critical gap in publicly available, high-quality data. Leveraging UMTID, we conduct a systematic, reproducible evaluation of state-of-the-art detectors—including YOLOv8, YOLOv10, YOLO11, and RT-DETR—establishing standardized performance baselines. Experimental results demonstrate the superior robustness of thermal imaging under extreme underground conditions (e.g., total darkness, occlusion), achieving an mAP of 72.3%, substantially outperforming visible-light-based approaches. This work provides foundational data and validated algorithms to advance autonomous perception capabilities for mine rescue robots.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Underground mining operations face significant safety challenges that make emergency response capabilities crucial. While robots have shown promise in assisting with search and rescue operations, their effectiveness depends on reliable miner detection capabilities. Deep learning algorithms offer potential solutions for automated miner detection, but require comprehensive training datasets, which are currently lacking for underground mining environments. This paper presents a novel thermal imaging dataset specifically designed to enable the development and validation of miner detection systems for potential emergency applications. We systematically captured thermal imagery of various mining activities and scenarios to create a robust foundation for detection algorithms. To establish baseline performance metrics, we evaluated several state-of-the-art object detection algorithms including YOLOv8, YOLOv10, YOLO11, and RT-DETR on our dataset. While not exhaustive of all possible emergency situations, this dataset serves as a crucial first step toward developing reliable thermal-based miner detection systems that could eventually be deployed in real emergency scenarios. This work demonstrates the feasibility of using thermal imaging for miner detection and establishes a foundation for future research in this critical safety application.

Problem

• Lack of comprehensive datasets for underground miner detection
• Need for reliable thermal-based miner detection systems
• Evaluating object detection algorithms for emergency scenarios

Innovation

• Novel thermal imaging dataset for miner detection
• Evaluated YOLOv8, YOLOv10, YOLO11, RT-DETR algorithms
• Thermal imaging feasibility for emergency scenarios