|本期目录/Table of Contents|

[1]王成昌,刘华,陈月,等.基于多特征无量纲向量的图像型管廊火灾探测技术研究*[J].中国安全生产科学技术,2023,19(4):14-20.[doi:10.11731/j.issn.1673-193x.2023.04.002]
　WANG Chengchang,LIU Hua,CHEN Yue,et al.Study on image-type fire detection technology of pipe gallery based on multi-feature dimensionless vector[J].Journal of Safety Science and Technology,2023,19(4):14-20.[doi:10.11731/j.issn.1673-193x.2023.04.002]
点击复制

基于多特征无量纲向量的图像型管廊火灾探测技术研究*

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
19
期数:
2023年4期
页码:
14-20
栏目:
学术论著
出版日期:
2023-04-30

文章信息/Info

Title:
Study on image-type fire detection technology of pipe gallery based on multi-feature dimensionless vector
文章编号:
1673-193X(2023)-04-0014-07
作者:
王成昌; 刘华; 陈月; 李享
（1.武汉理工大学 安全科学与应急管理学院，湖北 武汉 430070；
2.武汉纺织大学 环境工程学院，湖北 武汉 430299）
Author(s):
WANG Chengchang;  LIU Hua;  CHEN Yue;  LI Xiang
(1.School of Safety Science and Emergency Management,Wuhan University of Technology,Wuhan Hubei 430070,China;
2.School of Environmental Engineering,Wuhan Textile University,Wuhan Hubei 430299,China)
关键词:
多特征融合; 无量纲; 背景差分法; 管廊火灾; 图像型火灾探测
Keywords:
multi-feature fusion;  dimensionless;  background difference method;  pipe gallery fire;  image-type fire detection
分类号:
X932
DOI:
10.11731/j.issn.1673-193x.2023.04.002
文献标志码:
A
摘要:
针对管廊火灾检测中背景干扰与小目标检测困难的问题，提出采用无量纲特征参数描述图像型火焰动静态特征的火灾检测方法。利用背景差分法分割火焰目标区域，分析火焰发展特性，提取、计算角二阶矩变化率（A*ASM）、对比度变化率（C*CON）、自相关度变化率（C*COR）、面积变化率（S*）等无量纲特征参数；拟合无量纲特征融合向量T*并在SVM模型中检验其可靠性。研究结果表明：通过对不同可燃物在不同空间环境下的燃烧视频随机200帧序列进行检测，发现提出的无量纲特征参数具有“稳定”的变化范围，对火灾火焰与一般火焰的识别率最高可达97.5％，98.0％，受监测空间环境的影响较小。研究结果可为图像型火焰特征提取提供借鉴。
Abstract:
Aiming at the problems of background interference and small target detection in the pipe gallery fire detection,a fire detection method using dimensionless feature parameters to describe the dynamic and static characteristics of image-type flame was proposed.The background difference method was used to segment the flame target area,and the flame development characteristics were analyzed.The dimensionless feature parameters were extracted and calculated,such as the angular second moment change rate (A*ASM),contrast change rate (C*CON),autocorrelation change rate (C*COR) and area change rate (S*),then the dimensionless feature fusion vector T* was fitted,and its reliability was tested in the SVM model.The results showed that through detecting the random 200 frame sequences of combustion video of different combustibles in different spatial environments,it was found that the proposed dimensionless feature parameters had a stable range of variation,and the accuracy of fire detection was as high as 98%,which was less affected by the monitoring spatial environment.The research results can provide reference for the image-type flame feature extraction.

参考文献/References:

［1］杨娜娟,王慧琴,马宗方.基于支持向量机的图像型火灾探测算法［J］.计算机应用,2010,30(4):1129-1131，1140. YANG Najuan,WANG Huiqin,MA Zongfang.Image fire detection algorithm basedon support vector machine［J］.Computer Applications,2010,30(4):1129-1131，1140.
［2］金华彪.基于数字图像处理的火灾探测技术［J］.消防科学与技术,2002，7(3):46-47. JIN Huabiao.Fire detection technology based on digital image processing［J］.Fire Science and Technology,2002,7(3):46-47.
［3］段锁林,毛丹.关于火灾火焰图像检测算法研究［J］.计算机仿真,2016,33(2):393-398. DUAN Suolin,MAO Dan.Research on fire flame image detection algorithm［J］.Computer Simulation,2016,33(2):393-398.
［4］胡勤,陈琛,刘敏.早期火灾发生区域检测算法研究［J］.消防科学与技术,2013,32(12):1391-1393. HU Qin,CHEN Chen,LIU Min.Research on early fire occurrence area detection algorithm［J］.Fire Science and Technology,2013,32(12):1391-1393.
［5］袁非牛,廖光煊,张永明,等.计算机视觉火灾探测中的特征提取［J］.中国科学技术大学学报,2006,5(1):39-43. YUAN Feiniu,LIAO Guangxuan,ZHANG Yongming,et al.Feature extraction in fire detection with computer vision［J］.Journal of University of Science and Technology of China,2006,5(1):39-43.
［6］陈晓娟,卜乐平,李其修.基于图像处理的明火火灾探测研究［J］.海军工程大学学报,2007(3):6-11. CHEN Xiaojuan,BU Leping,LI Qixiu.Research on open flame fire detection based on image processing［J］.Journal of Naval University of Engineering,2007,7(3):6-11.
［7］白卫东,严建华,马增益,等.基于支持向量机的火焰状态识别方法［J］.动力工程,2004(4):548-551. BAI Weidong,YAN Jianhua,MA Zengyi,et al.Flame state identification method based on support vector machine［J］.Power Engineering,2004(4):548-551.
［8］KOERICH B,PAULOH.A probabilistic approach for vision-based fire detection in videos［J］.IEEE Transactions on Circuits & Systems for Video Technology,2010,20(5):721-731.
［9］黄正宇,缪小平,芮挺.基于图像的火焰检测中无量纲动态特征研究［J］.计算机应用,2012,32(7):1894-1898. HUANG Zhengyu,MIAO Xiaoping,RUI Ting.Research on dimensionless dynamic features in flame detection based on image［J］.Computer Applications,2012,32(7):1894-1898.
［10］ZHENG Q,CHEN Y.Feature pyramid of bidirectional stepped concatenation for small object detection［J］.Multimedia Tools and Applications,2021(11):1-23.
［11］QU J S,SU C,ZHANG Z,et al.Dilated convolution and feature fusion SSD network for small object detection in remote sensing images［J］.IEEE Access,2020,11(99):1-10.
［12］张艳,张明路,吕晓玲,等.深度学习小目标检测算法研究综述［J］.计算机工程与应用,2022,58(15):1-17. ZHANG Yan,ZHANG Minglu,LYU Xiaoling,et al.Review of deep learning small target detection algorithm［J］.Computer Engineering and Application,2022,58(15):1-17.
［13］刘洋,战荫伟.基于深度学习的小目标检测算法综述［J］.计算机工程与应用,2021,57(2):37-48. LIU Yang,ZHAN Yinwei.Review of small target detection algorithms based on deep learning［J］.Computer Engineering and Applications,2021,57(2):37-48.
［14］GIRSHICK R,DONAHUE J,DARRELL T,et al.Rich feature hierarchies for accurate object detection and semantic segmentation［C］//IEEE Conference on Computer Vision and Pattern Recognition.Columbus:2014.
［15］REN S,HE K,GIRSHICK R,et al.Faster R-CNN:towards realtime object detection with region proposal networks［J］.IEEE Transactions on Pattern Analysis & Machine Intelligence,2017,39(6):1137-1149.
［16］LIN T Y,DOLLAR P,GIRSHICK R,et al.Feature pyramid networks for object detection［J］.2017 IEEE Conference on Computer Vision and Pattern Recognition,2017(4):936-944.
［17］许德刚,王露,李凡.深度学习的典型目标检测算法研究综述［J］.计算机工程与应用,2021,57(8):10-25. XU Degang,WANG Lu,LI Fan.Review of typical object detection algorithms for deep learning［J］.Computer Engineering and Applications,2021,57(8):10-25.
［18］陈长友,杨健晟.面向视频图像的烟雾检测算法综述［J］.激光与光电子学进展,2021,58(4):29-43. CHEN Changyou,YANG Jiansheng.Review of smoke detection algorithms for video images［J］.Progress in Laser & Optoelectronics,2021,58(4):29-43.
［19］荣建忠.基于多特征的火焰图像探测研究及实现［D］.合肥:中国科学技术大学,2012.
［20］屈晶晶,辛云宏.连续帧间差分与背景差分相融合的运动目标检测方法［J］.光子学报,2014,43(7):219-226. QU Jingjing,XIN Yunhong.Moving target detection method for fusion of continuous interframe difference and background difference［J］.Acta Photonica Sinica,2014,43(7):219-226.
［21］魏志强,纪筱鹏,冯业伟.基于自适应背景图像更新的运动目标检测方法［J］.电子学报,2005(12):2261-2264. WEI Zhiqiang,JI Xiaopeng,FENG Yewei.Moving object detection method based on adaptive background image update［J］.Acta Electronica Sinica,2005(12):2261-2264.
［22］张娟,毛晓波,陈铁军.运动目标跟踪算法研究综述［J］.计算机应用研究,2009,26(12):4407-4410. ZHANG Juan,MAO Xiaobo,CHEN Tiejun.Review of moving target tracking algorithm［J］.Application Research of Computers,2009,26(12):4407-4410.
［23］阳婷,官洪运.基于帧间高频能量和相关性的烟雾检测算法研究［J］.微型机与应用,2015,34(17):32-35. YANG Ting,GUAN Hongyun.Research on smoke detection algorithm based on high-frequency energy and correlation between frames［J］.Microcomputer and Application,2015,34(17):32-35.
［24］CHEN Z,ELLIS T.A selfadaptive Gaussian mixture model［J］.Computer Vision & Image Understanding,2014,122(5):35-46.
［25］ZHAO J H,ZHANG Z,HAN S Z,et al.SVM based forest fire detectionusing static and dynamic features［J］.Computer Science and Information Systems,2011,8(3):21-38.

相似文献/References:

[1]王冠宁,陈涛,邢菲,等.监控完全遮挡场景下火焰特征分析方法*[J].中国安全生产科学技术,2021,17(9):13.[doi:10.11731/j.issn.1673-193x.2021.09.002]
　WANG Guanning,CHEN Tao,XING Fei,et al.Analysis method of flame characteristic in scene of completely blocked surveillance[J].Journal of Safety Science and Technology,2021,17(4):13.[doi:10.11731/j.issn.1673-193x.2021.09.002]

备注/Memo

备注/Memo:
收稿日期： 2022-05-21
* 基金项目： 国家自然科学基金项目( 52074202)
作者简介： 王成昌，博士，副教授，主要研究方向为经济技术与投资评价、企业战略管理以及突发事件应急管理。
通信作者： 陈月，博士，讲师，主要研究方向为粉尘爆炸及生物质基抑爆技术。
更新日期/Last Update: 2023-05-11