Research on early fire source identification and anti-interference methods in mines based on dual-spectrum imaging technology
-
摘要:
现有基于图像分析的矿井外因火灾监测方法受矿井环境复杂、干扰源影响较大，单模态方法易将光源误判为火源，多模态方法没有利用温度信息进行火源判定，且在粉尘条件下这两种方法的识别精度较低。针对上述问题，提出一种基于双光谱成像技术的矿井早期火源识别及抗干扰方法。首先采用YOLOv10模型对可见光图像进行实时火源检测，利用红外热成像获取温度分布数据，然后通过Canny边缘检测与图像二值化预处理，消除可见光与红外图像的成像差异，最后采用pHash算法计算可见光与红外图像边缘哈希值的海明距离，并标定阈值（海明距离≤25），判定是否为同一火源，从而有效区分火源与干扰源。实验结果表明：在无粉尘无干扰源工况下，基于双光谱成像技术的矿井早期火源识别及抗干扰方法的准确率达98%，召回率为94%，优于单模态的YOLOv10（准确率为97%，召回率为86%）；在粉尘干扰条件下，粉尘覆盖摄像头表面33%时，该方法的准确率和召回率分别为85%，80%，粉尘覆盖摄像头表面66%时, 准确率和召回率分别为70%，65%，优于单模态和多模态方法。
Abstract:Existing image analysis-based methods for exogenous mine fire detection are affected by complex mining environments and interference sources. Single-modal methods tend to misjudge light sources as fire sources, while multi-modal methods fail to utilize temperature information for fire source identification. Additionally, both methods have low identification accuracy under dust conditions. To address the above issues, an early fire source identification and anti-interference method for mines based on dual-spectrum imaging technology was proposed. First, the YOLOv10 model was used for real-time fire source detection on visible light images, and infrared thermal imaging was employed to obtain temperature distribution data. Then, Canny edge detection and image binarization preprocessing were applied to eliminate imaging differences between visible light and infrared images. Finally, the pHash algorithm was used to calculate the Hamming distance of the edge hash values between visible light and infrared images, and a threshold (Hamming distance≤25) was set to determine whether they represented the same fire source, thus effectively distinguishing fire sources from interference sources. The experimental results showed that under conditions without dust or interference sources, the accuracy of the early fire source detection and anti-interference method based on dual-spectrum imaging technology reached 98%, with a recall rate of 94%, outperforming the single-modal YOLOv10 (accuracy 97%, recall rate 86%). Under dust interference conditions, when 33% of the camera surface was covered by dust, the accuracy and recall rates were 85% and 80%, respectively. When 66% of the camera surface was covered by dust, the accuracy the recall rate were 70% and 65%, which were superior to both single-modal and multi-modal methods.
-
-
![]()
图 1 双光谱成像方法流程
Figure 1. Process of early fire source identification and anti-interference method for mines based on dual spectrum imaging
![]()
图 10 二值化前后边缘提取对比
Figure 10. Comparison of edge extraction before and after binarization
表 1 无粉尘无干扰源下火源监测方法对比结果
Table 1 Comparison of fire source monitoring method under conditions without dust or interference
% 方法 准确率 召回率 YOLOv10 0.97 0.86 双光谱成像方法 0.98 0.94 
下载: 导出CSV
表 2 粉尘条件下火源识别方法对比结果
Table 2 Comparison of fire source monitoring method under dust conditions
% 方法 粉尘浓度33% 粉尘浓度 66% 粉尘浓度 99% 准确率 召回率 准确率 召回率 准确率 召回率 YOLOv10 0.70 0.58 0.40 0.25 0 0 多模态图像融合 0.70 0.72 0.50 0.40 0 0 双光谱成像方法 0.85 0.80 0.70 0.65 0 0
下载: 导出CSV
-
[1] 邓军,李鑫,王凯,等. 矿井火灾智能监测预警技术近20年研究进展及展望[J]. 煤炭科学技术,2024,52(1):154-177. DENG Jun,LI Xin,WANG Kai,et al. Research progress and prospect of mine fire intelligent monitoring and early warning technology in recent 20 years[J]. Coal Science and Technology,2024,52(1):154-177.
[2] 邓军,张琦,陈炜乐,等. 矿井煤自燃灾害监测预警技术及发展趋势[J]. 煤矿安全,2024,55(3):99-110. DENG Jun,ZHANG Qi,CNEN Weile,et al. Coal spontaneous combustion disaster monitoring and early warning technologies and development trend for coal mines[J]. Safety in Coal Mines,2024,55(3):99-110.
[3] 孙继平,李小伟,徐旭,等. 矿井电火花及热动力灾害紫外图像感知方法研究[J]. 工矿自动化,2022,48(4):1-4,95. SUN Jiping,LI Xiaowei,XU Xu,et al. Research on ultraviolet image perception method of mine electric spark and thermal power disaster[J]. Journal of Mine Automation,2022,48(4):1-4,95.
[4] 王远声,景国勋,郭绍帅,等. 煤矿瓦斯爆炸灾害的复杂网络分析与断链减灾措施研究[J/OL]. 安全与环境学报:1-11[2024-11-24]. https://doi.org/10.13637/j.issn.1009-6094.2024.0569. WAGN Yuansheng,JING Guoxun,GUO Shaoshuai,et al. Complex network analysis for disaster chain evaluation and mitigation of coal mine gas explosions[J/OL]. Journal of Safety and Environment:1-11[2024-11-24]. https://doi.org/10.13637/j.issn.1009-6094.2024.0569.
[5] 张洪亮. 基于虚拟现实技术的煤矿胶带火灾模拟系统[J]. 煤矿安全,2014,45(4):128-131. ZHANG Hongliang. Coal mine belt fire simulation system based on virtual reality technology[J]. Safety in Coal Mines,2014,45(4):128-131.
[6] 赵文军. 矿井火灾爆炸危险性评估与防控技术研究[J]. 能源与节能,2025(2):179-181. DOI: 10.3969/j.issn.2095-0802.2025.02.048 ZHAO Wenjun. Mine fire and explosion risk assessment and prevention and control technologies[J]. Energy and Energy Conservation,2025(2):179-181. DOI: 10.3969/j.issn.2095-0802.2025.02.048
[7] 徐宏宇,续婷. 一种基于颜色和纹理的优化SVM火灾识别方法[J]. 沈阳航空航天大学学报,2021,38(4):54-60. DOI: 10.3969/j.issn.2095-1248.2021.04.007 XU Hongyu,XU Ting. A color/texture-based improved SVM for fire recognition[J]. Journal of Shenyang Aerospace University,2021,38(4):54-60. DOI: 10.3969/j.issn.2095-1248.2021.04.007
[8] 熊昊,李伟. 基于SVM的视频火焰检测算法[J]. 传感器与微系统,2020,39(1):143-145,149. XIONG Hao,LI Wei. Video flame detection algorithm based on SVM[J]. Transducer and Microsystem Technologies,2020,39(1):143-145,149.
[9] 王亚,张宝峰. 基于显著性检测的红外森林火灾监测系统[J]. 消防科学与技术,2018,37(12):1700-1703. DOI: 10.3969/j.issn.1009-0029.2018.12.029 WANG Ya,ZHANG Baofeng. Infrared forest fire monitoring system based on saliency detection[J]. Fire Science and Technology,2018,37(12):1700-1703. DOI: 10.3969/j.issn.1009-0029.2018.12.029
[10] 王思嘉,裴海龙. 基于火焰图像红外动态特征的早期火灾识别算法[J]. 现代电子技术,2010,33(8):104-105,110. DOI: 10.3969/j.issn.1004-373X.2010.08.033 WANG Sijia,PEI Hailong. Algorithm for early fire recognition based on infrared dynamic characteristics of flame images[J]. Modern Electronics Technique,2010,33(8):104-105,110. DOI: 10.3969/j.issn.1004-373X.2010.08.033
[11] 刘培江,董辉,宋子刚,等. 基于视频图像处理技术的火焰识别算法[J]. 热能动力工程,2021,36(3):64-71. LIU Peijiang,DONG Hui,SONG Zigang,et al. Flame recognition algorithm based on video and image processing technology[J]. Journal of Engineering for Thermal Energy and Power,2021,36(3):64-71.
[12] 孙继平,崔佳伟. 矿井外因火灾感知方法[J]. 工矿自动化,2021,47(4):1-5,38. SUN Jiping,CUI Jiawei. Mine external fire sensing method[J]. Industry and Mine Automation,2021,47(4):1-5,38.
[13] 孙继平,李月. 基于双目视觉的矿井外因火灾感知与定位方法[J]. 工矿自动化,2021,47(6):12-16,78. SUN Jiping,LI Yue. Binocular vision-based perception and positioning method of mine external fire[J]. Industry and Mine Automation,2021,47(6):12-16,78.
[14] 韩斌,吴一全,宋昱. 基于改进CV模型的煤矿井下早期火灾图像分割[J]. 煤炭学报,2017,42(6):1620-1627. HAN Bin,WU Yiquan,SONG Yu. Segmentation of early fire image of mine based on improved CV model[J]. Journal of China Coal Society,2017,42(6):1620-1627.
[15] 梁运涛,王伟. 矿井外因火灾监测预警与智能防控技术[J/OL]. 矿业安全与环保:1-8[2024-11-17]. http://kns.cnki.net/kcms/detail/50.1062.TD.20241022.1004.002.html. LIANG Yuntao,WANG Wei. Mine exogenous fire monitoring and early warning and intelligent prevention and controltechnology[J/OL]. Mining Safety & Environmental Protection:1-8[2024-11-17]. http://kns.cnki.net/kcms/detail/50.1062.TD.20241022.1004.002.html.
[16] 范伟强. 矿井外因火灾双光谱图像监测方法研究[D]. 北京:中国矿业大学(北京),2022. FAN Weiqiang. Study on dual-spectral image monitoring method of mine external fire[D]. Beijing:China University of Mining & Technology-Beijing,2022.
[17] 孙继平,范伟强. 矿井红外热成像远距离测温误差分析与精确测温方法[J]. 煤炭学报,2022,47(4):1709-1722. SUN Jiping,FAN Weiqiang. Error analysis and accurate temperature measurement method of infrared thermal imaging long-distance temperature measurement in underground mine[J]. Journal of China Coal Society,2022,47(4):1709-1722.
[18] 李益明,卜雄洙,沈樾. 基于红外传感器和图像识别的复合式火焰检测技术研究[J]. 仪表技术,2024(4):39-43,59. LI Yiming,BU Xiongzhu,SHEN Yue. Research on composite flame detection technology based on infrared sensors and image recognition[J]. Instrumentation Technology,2024(4):39-43, 59.
[19] 孙继平,李小伟. 基于图像内凹度的矿井外因火灾识别及抗干扰方法[J]. 煤炭学报,2024,49(7):3253-3264. SUN Jiping,LI Xiaowei. Mine external fire recognition and anti-interference method based on the internal concavity of image[J]. Journal of China Coal Society,2024,49(7):3253-3264 .
[20] 刘汝琪. 基于多模态图像的火灾检测算法研究[D]. 西安:中国科学院大学(中国科学院西安光学精密机械研究所),2022. LIU Ruqi. Research on fire detection algorithms based on multimodalImages[D]. Xi'an:Xi'an Institute of Optics and Precision Mechanics,Chinese Academy of Sciences,2022.
[21] 金政北,金贝贝,宋晓辉,等. 改进YOLOv10算法及其在路面坑洼检测中的应用[J/OL]. 计算机应用与软件:1-8[2024-11-24]. http://kns.cnki.net/kcms/detail/31.1260.tp.20250307.0824.002.html. JIN Zhengbei,JIN Beibei,SONG Xiaohui,et al. Improved YOLOv10 algorithm and its application on pothole detection[J/OL]. Computer Applications and Software:1-8[2024-11-24]. http://kns.cnki.net/kcms/detail/31.1260.tp.20250307.0824.002.html.
-
期刊类型引用(5)
1. 陈盈宜，潘丽敏，叶勇，李君，黄光文. 基于SVM算法的超声波速度-土壤含水率估计模型. 华中农业大学学报. 2024(02): 247-253 . 
百度学术
2. 韩世礼，肖健，柳位. 机器学习在地球物理勘探中铀矿资源勘查的应用研究进展. 铀矿地质. 2024(03): 555-564 . 百度学术
3. 魏礼鹏，鹿伟强，于铄航，陈雯雅，张珂. 基于SA-PSO的煤矿小功率电机故障诊断研究. 煤矿机械. 2024(07): 174-176 . 百度学术
4. 宋成林，黄晓冉，邢帅，芦楠楠. 基于隐半马尔可夫模型的微震信号分割方法. 中国科技论文. 2024(08): 868-876 . 百度学术
5. 王志坚，芦盛亮，李钢，查颖，蔡宏伟，韩大永. 基于现场矿压显现的微震预警指标研究. 煤炭与化工. 2023(06): 82-84+88 . 百度学术
其他类型引用(1)
计量
- 文章访问数: 20
- HTML全文浏览量: 1
- PDF下载量: 2
- 被引次数: 6
