基于知识图谱的火灾图像识别研究

doi:10.16265/j.cnki.issn1003-3033.2023.10.2088

摘要/Abstract

摘要：

为了全面分析图像型火灾识别技术的发展趋势和研究动态,更准确地为火灾探测领域科研工作提供研究方向,运用Web of Science已有文献数据和科学知识图谱软件以及Python-matplotlib库,定量分析国际火灾图像研究的发文时间、发文作者、发文机构、高被引文献等特征,从主题、关键词、摘要入手分析当前研究热点与前沿趋势动态。结果表明:国际火灾图像识别研究成果数量整体上呈现出波浪式上升趋势;欧美国家在火灾图像识别领域的研究较为深入,而中国在该领域研究起步较晚;J Comp Neurol、Remote Sens Environ、Fire Safety J、J Geophys Res-Atmos为代表期刊形成共被引期刊合作网络;研究热点主要表现在火灾图像识别的深度学习模型、森林火灾图像及火灾影响、火灾遥感图像识别算法3个方面;研究前沿主要表现在基于深度学习的火灾烟雾探测,火烧迹地的森林植被覆盖与流失,煤矿、工业热源、电动汽车的火灾探测,阻燃性4个方面。

关键词: 火灾探测, 火灾图像识别, 知识图谱, 研究热点, 深度学习

Abstract:

In order to comprehensively analyze the development trend and research trends of image-based fire identification technology, and more accurately provide research direction for scientific research in the field of fire detection, the Web of Science existing literature data and scientific knowledge mapping software, Python-maplotlib library, etc., were used to quantitatively analyze the characteristics of international fire image research, such as the time of publication, the author, the organisation, and the highly cited articles. This paper started with an analysis of current research hotspots and frontier trends. The results show that the number of international fire image recognition research achievements shows a wavy upward trend. The research on fire image and its related fields in Europe and America is relatively deep, while that in China is relatively late. J Comp Neurol, Remote Sens Environment, Fire Safety J, J Geophys Res Atmos are representative journals that form a cooperative network of co-cited journals. The research focuses mainly on the deep learning model of fire image recognition, forest fire image and fire impact, and fire remote sensing image recognition algorithm. The research frontiers are mainly shown in four aspects: fire smoke detection based on deep learning, forest vegetation coverage and loss of burned areas, fire detection of coal mines, industrial heat sources, electric vehicles, and flame retardancy.

Key words: fire detection, fire images recognition, scientific knowledge graph, research hotspots, deep learning

李海, 孙鹏. 基于知识图谱的火灾图像识别研究[J]. 中国安全科学学报, 2023, 33(10): 147-159.

LI Hai, SUN Peng. Research on fire image recognition based on scientific knowledge graph[J]. China Safety Science Journal, 2023, 33(10): 147-159.

图/表 23

图1

图2

图3

图4

表1

图5

图6

图7

表2

图8

图9

图10

图11

表3

图12

图13

表4

图14

图15

图16

图17

图18

表5

参考文献 30

[1]	国务院安全生产委员会. “十四五”国家消防工作规划[Z].2022-02-12.
[2]	国务院公安部. 关于全面推进“智慧消防”建设的指导意见[Z].2017-10-10.
[3]	国务院应急管理部消防救援局. 上半年全国共处置各类警情96.2万起[OL].(2022-07-23). https://www.chinanews.com.cn/gn/2022/07-23/9810180.shtml.
[4]	刘全义, 朱博, 邓力, 等. 基于机器学习的双参数火灾探测方法[J]. 中国安全科学学报, 2022, 32(5):90-96. doi: 10.16265/j.cnki.issn1003-3033.2022.05.0874
	LIU Quanyi, ZHU Bo, DENG Li, et al. Double parameters fire detection based on machine learning[J]. China Safety Science Journal, 2022, 32(5):90-96. doi: 10.16265/j.cnki.issn1003-3033.2022.05.0874
[5]	何永勃, 董玉珊, 雷建, 等. 飞机货舱火灾多传感器探测方法研究[J]. 中国安全科学学报, 2018, 28(5):74-79. doi: 10.16265/j.cnki.issn1003-3033.2018.05.013
	HE Yongbo, DONG Yushan, LEI Jian, et al. Research on multi-sensor method for detecting aircraft cargo fire[J]. China Safety Science Journal, 2018, 28 (5): 74-79. doi: 10.16265/j.cnki.issn1003-3033.2018.05.013
[6]	陈悦, 陈超美, 刘则渊, 等. CiteSpace知识图谱的方法论功能[J]. 科学学研究, 2015, 33(2):242-253.
	CHEN Yue, CHEN Chaomei, LIU Zeyuan, et al. The methodological function of SiteSpace knowledge graph[J]. Scientific Research, 2015, 33 (2): 242-253.
[7]	杨倩. 常见文献计量学工具的分析功能比较研究[J]. 情报探索, 2021(10): 87-93.
	YANG Qian. Comparative research on the analysis function of common bibliometrics tools[J]. Information Research, 2021 (10): 87-93.
[8]	SCHROEDER W, OLIVA P, GIGLIO L, et al. The new VIIRS 375 m active fire detection data product: algorithm description and initial assessment[J]. Remote Sensing of Environment, 2014, 143: 85-96. doi: 10.1016/j.rse.2013.12.008
[9]	GIGLIO L, SCHROEDER W, JUSTICE C O. The collection 6 MODIS active fire detection algorithm and fire products[J]. Remote Sensing of Environment, 2016, 178: 31-41. doi: 10.1016/j.rse.2016.02.054 pmid: 30158718
[10]	ZHENG Yanning, XU Xiaoyang, LIU Zhihui. Study on the method of identifying research fronts based on keywords co-occurrence[J]. Library and Information Service, 2016, 60(4):85-92. doi: 10.13266/j.issn.0252-3116.2016.04.012
[11]	GU Ke, XIA Zhifang, QIAO Junfei, et al. Deep dual-channel neural network for image-based smoke detection[J]. IEEE Transactions on Multimedia, 2020, 22(22):311-323. doi: 10.1109/TMM.6046
[12]	PENG Yingshu, WANG Yi. Real-time forest smoke detection using hand-designed features and deep learning[J]. Computers and Electronics in Agriculture, 2019,167:DOI:10.1016/j.compag.2019.105029.
[13]	ZHAO Yi, MA Jiale, LI Xiahui, et al. Saliency detection and deep learning based wildfire identification in UAV imagery[J]. Sensors, 2018, 18(3): DOI:10.3390/S18030712.
[14]	PUNDIR A S, RAMAN B. Dual deep learning model for image based smoke detection[J]. Fire Technology, 2020, 55(6):2419-2442. doi: 10.1007/s10694-019-00872-2
[15]	LARSEN A, HANIGAN I, REICH B J, et al. A deep learning approach to identify smoke plumes in satellite imagery in near-real time for health risk communication[J]. Journal of Exposure Science and Environmental Epidemiology, 2020, 31(1):170-176. doi: 10.1038/s41370-020-0246-y
[16]	MUHAMMAD K, AHMAD J, MEHMOOD I, et al. Convolutional neural networks based fire detection in surveillance videos[J]. IEEE Access, 2018, 6:18174-18183. doi: 10.1109/ACCESS.2018.2812835
[17]	LIMA P D, MARFURT. Convolutional neural network for remote sensing scene classification: transfer learning analysis[J]. Remote Sensing, 2019, 12(1): DOI:10.3390/rs12010086.
[18]	BAN Yifang, ZHANG Puzhao, NASCETTI A, et al. Near real time wildfire progression monitoring with sentinel-1 SAR time series and deep learning[J]. Scientific Reports, 2020, 10(1):DOI:10.1038/s41598-019-56967-x.
[19]	NAMOZOV A, CHO Yi. An efficient deep learning algorithm for fire and smoke detection with limited data[J]. Advances in Electrical and Computer Engineering, 2018, 18(4) :121-128.
[20]	SAEED F, PAUL A, KARTHIGAIKUMAR P, et al. Convolutional neural network based early fire detection[J]. Multimedia Tools and Applications, 2020, 79: 9083-9099. doi: 10.1007/s11042-019-07785-w
[21]	MAO Wentao, WANG Wenpeng, DOU Zhi, et al. Fire recognition based on multi-channel convolutional neural network[J]. Fire Technology, 2018, 54(2):531-554. doi: 10.1007/s10694-017-0695-6
[22]	PARK M, DAI Q T, JUNG D, et al. Wildfire detection method using denseNet and CycleGAN data augmentation based remote camera imagery[J]. Remote Sensing, 2020, 12(22):DOI:10.3390/rs12223715.
[23]	YANG Hunjun, JANG H, KIM T, et al. Non-temporal lightweight fire detection network for intelligent surveillance systems[J]. IEEE Access, 2019, 7: 169 257-169 266.
[24]	KIM B, LEE J. A Video-based fire detection using deep learning models[J]. Applied Sciences-Basel, 2019, 9(14): DOI:10.3390/App9142862.
[25]	CAO Yichao, YANG Feng, TANG Qingfei, et al. An attention enhanced bidirectional LSTM for early forest fire smoke recognition[J]. IEEE Access, 2019, 7:154732-154742. doi: 10.1109/ACCESS.2019.2946712
[26]	XU Renjie, LIN Haifei, LU Kangjie, et al. A forest fire detection system based on ensemble learning[J]. Forestry, 2021, 12(2):DOI:10.3390/f12020217.
[27]	KHAN S, MUHAMMAD K, HUSSAIN T, et al. Deep smoke: deep learning model for smoke detection and segmentation in outdoor environments[J]. Expert Systems with Applications, 2021, 182(1):115-125.
[28]	BARMPOUTIS P, DIMITROPOULOS K, KAZA K, et al. Fire detection from images using faster R-CNN and multidimensional texture analysis[C]. ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2019: 8301-8305.
[29]	CHAOXIA Chenyu, SHANG Weiwei, ZHANG Fei. Information guided flame detection based on faster R-CNN[J]. IEEE Access, 2020, 8:58923-58 932.
[30]	ZHONG Zhen. WANG Minjuan, SHI Yukun, et al. A convolutional neural network-based flame detection method in video sequence[J]. Signal Image and Video Processing, 2018, 12(8): 1619-1627. doi: 10.1007/s11760-018-1319-4

排名	国家/地区	大洲	文献数量	引用量	篇均被引数
1	美国	北美洲	2 475	60 392	24.400 8
2	中国	亚洲	1 862	22 690	12.185 8
3	德国	欧洲	515	11 369	22.075 7
4	英国	欧洲	508	12 029	23.679 1
5	法国	欧洲	491	9 443	19.232 2
6	加拿大	北美洲	420	8 330	19.833 3
7	印度	亚洲	413	3 783	9.159 8
8	西班牙	欧洲	408	8 602	21.083 3
9	澳大利亚	大洋洲	389	8 739	22.465 3
10	意大利	欧洲	349	7 395	21.189 1
11	韩国	亚洲	333	4 552	13.669 7
12	日本	亚洲	329	3 848	11.696
13	巴西	南美洲	256	4 218	16.476 6
14	荷兰	欧洲	177	7 778	43.943 5
15	俄罗斯	欧洲	154	1 372	8.909 1
16	瑞士	欧洲	141	3 562	25.262 4
17	希腊	欧洲	120	1 987	16.558 3
18	台湾	亚洲	115	828	7.2
19	土耳其	亚、欧洲	114	981	8.605 3
20	比利时	欧洲	113	3 020	26.725 7

排名	作者	所在机构	被引数量/次
1	GIGLIO	University of Maryland	1 260
2	CHUVIECO	Universidad de Alcalá	782
3	ROY	South Dakota State University	623
4	CELIK	University of the Witwatersrand	399
5	VAN	Vrije Universiteit Amsterdam	386
6	TOREYIN	Istanbul Technical University	380
7	SCHROEDER	National Oceanic and Atmospheric Administration	378
8	WOOSTER	King's College London	345

主题1		主题2		主题3
主题	频次	主题	频次	主题	频次
模型	374	火灾	682	中分辨成像光谱仪	394
动力学	286	遥感	371	算法	269
火灾探测	273	野火	347	排放物	202
分类	265	气候变化	327	验证	172
神经元	199	植被	293	生物量	153
系统	191	陆地卫星	236	卫星	144
性能	189	森林	212	毁林	118
深度学习	188	时间序列	164	燃烧区域	103
温度	182	模式	154	森林砍伐变异性	99
图像处理	178	植被指数	136	碳	90

基于Title前沿分析		基于Kerword前沿分析		基于Abstact前沿分析
vegetation cover	植被覆盖	deep learning	(1)深度学习	smoke detection	(1)烟雾探测
coal fire	(3)煤矿火灾	remote sensing	遥感	burned area	(2)火烧迹地
smoke detection	(1)烟雾探测	burned area	(2)火烧迹地	areaforest-cover loss	森林植被流失
flame retardancy	(4)阻燃性	neuron	神经元	severity map	严重程度图
optical depth	光学深度	flame retardancy	(4)阻燃性	industrial heat source	(3)工业热源