Abstract

Underground mine tunnel fires pose serious threats to personnel safety and infrastructure integrity, yet their detection and characterization remain challenging due to complex ventilation environments and limited visibility. This study combines full-scale fire experiments with advanced computer vision analysis to investigate the influence of fire size and ventilation on thermal, chemical, and visual signatures of mine fires. Six full-scall controlled experiments were conducted using two fire pan sizes under three ventilation conditions. Fuel mass loss, temperature distribution, and gas concentrations were systematically recorded using thermocouples, gas sensors, and visual/thermal imaging. Results show that mass loss rate is primarily governed by fire size, while ventilation has only minor effects within the tested ventilation range. Temperature and gas measurements reveal distinct patterns: small pans exhibit moderate CO2 accumulation and ventilation-dependent attenuation, whereas large pans produce higher but short-lived peaks with negligible ventilation influence. Multi-camera flame imaging demonstrates that flame color and intermittency are strongly affected by smoke concentration, distance, and airflow. To support intelligent fire monitoring, knowledge-based features, including color, texture, and morphology, were extracted from multi-view camera images. These features enable high-accuracy classification of fire size and ventilation conditions, with multi-view fusion further enhancing stability and reducing errors (98.8% for fire size and 98.3% for ventilation). The findings provide quantitative insights into flame behavior under different operational scenarios and demonstrate the potential of combining experimental data with knowledge-based image analysis for improved fire detection and safety management in underground mines.

Department(s)

Mining Engineering

Publication Status

Full Text Access

Keywords and Phrases

Fire detection; Knowledge-based features; Multi-camera fire monitoring; Underground mine tunnel fire

International Standard Serial Number (ISSN)

0886-7798

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2026 Elsevier, All rights reserved.

Publication Date

01 Aug 2026

Download

Full Text Link

Share

 
COinS