[1]戴 静,严云洋,范 勇,等.基于BEMD和SVM的火焰检测算法[J].常州大学学报(自然科学版),2017,(02):71-77.[doi:10.3969/j.issn.2095-0411.2017.02.013]
　DAI Jing,YAN Yunyang,FAN Yong,et al.Fire Detection Based on Bidimensional Empirical Mode Decomposition and Support Vector Machine[J].Journal of Changzhou University(Natural Science Edition),2017,(02):71-77.[doi:10.3969/j.issn.2095-0411.2017.02.013]
点击复制

基于BEMD和SVM的火焰检测算法()
分享到：

常州大学学报(自然科学版)[ISSN:2095-0411/CN:32-1822/N]

卷:
期数:
2017年02期
页码:
71-77
栏目:
计算机与信息工程
出版日期:
2017-03-28

文章信息/Info

Title:
Fire Detection Based on Bidimensional Empirical Mode Decomposition and Support Vector Machine
作者:
戴 静1; 2; 严云洋1; 2; 范 勇1; 高尚兵2; 周静波 2
1. 西南科技大学 计算机科学与技术学院,四川 绵阳 621010; 2. 淮阴工学院 计算机与软件工程学院,江苏 淮安 223003
Author(s):
DAI Jing1; 2;  YAN Yunyang1; 2;  FAN Yong1;  GAO Shangbing2;  ZHOU Jingbo2
1. School of Computer Science and Technology, Southwest University of Science and Technology, Mianyang 621010,China; 2. Faculty of Computer & Software Engineering, Huaiyin Institute of Technology, Huaian 223003,China
关键词:
二维经验模式;  局部二值模式;  火焰检测;  纹理特征;  支持向量机
Keywords:
bidimensional empirical mode decomposition;  local binary patterns;  flame detection;  texture feature;  support vector machine
分类号:
TP 391.41
DOI:
10.3969/j.issn.2095-0411.2017.02.013
文献标志码:
A
摘要:
为提高火焰检测的准确性,提出了一种采用二维经验模式(BEMD)和支持向量机(SVM)的火焰检测算法。首先基于累积差分法检测运动目标,根据Ohta颜色空间找出图像中具有火焰颜色的疑似区域; 其次将疑似区域图像经过BEMD分解,结合局部二值模式(LBP)对所提取到的固有模态函数(IMF)图像进行纹理特征提取; 最后将提取的纹理特征结合圆形度、矩形度、重心高度输入到SVM里面进行火焰的判别。实验结果表明该方法具有较高的火焰检测率以及较低的误检率。
Abstract:
In order to improve the accuracy of fire detection,a fire detection algorithm based on Bidimensional Empirical Mode Decomposition(BEMD)and Support Vector Machine(SVM)is proposed. Firstly,candidate fire regions were detected based on the accumulative difference method for detecting moving targets and Ohta color space with color model of flame. Secondly,a new method combining the bidimensional empirical mode decomposition(BEMD)with local binary pattern(LBP)is proposed for texture image classification. The LBP is used to extract the features of a series of various intrinsic mode functions(IMFS)images and residual images,which are decomposed by bidimensional empirical mode from the image. Finally, the roundness, rectangle degree, height of center of gravity, texture features are input into the SVM classification. The experimental results show that this method has high flame detection rate, low false alarm rate.

参考文献/References:

[1]李正周, 方朝阳, 顾园山, 等. 基于无线多传感器信息融合的火灾检测系统[J]. 数据采集与处理, 2014,29(5):694-699.
[2]王文豪, 陈晓兵, 刘金岭. 基于连通区域和SVM特征融合的火灾检测[J]. 计算机仿真, 2014, 31(1):383-387.
[3]YANG X, WANG J, HE S. A SVM approach for vessel fire detection based on image processing[C]//Modelling, identification & control(ICMIC). Proceedings of International Conference on IEEE. Wuhan: [s.n.], 2012:150-153.
[4]ZHAO J, ZHANG Z, HAN S, et al. SVM based forest fire detection using static and dynamic features[J]. Computer Science and Information Systems, 2011, 8(3):821-841.
[5]李红娣,袁非牛. 采用金字塔纹理和边缘特征的图像烟雾检测[J]. 中国图象图形学报,2015(6):772-780.
[6]周书仁,殷建平. 基于Haar特性的LBP纹理特征[J]. 软件学报,2013(8):1909-1926.
[7]YUAN F N.Video-based smoke detection with histogram sequence of LBP and LBPV pyramids[J].Fire Safety Journal,2011,46:132-139.
[8]严云洋,吴茜茵,杜静,等.基于色彩和闪频特征的视频火焰检测[J].计算机科学与探索,2014,8(10):1271-1279
[9]范朝冬, 张英杰, 欧阳红林,等. 基于改进斜分Otsu法的回转窑火焰图像分割[J]. 自动化学报, 2014, 40(11):2480-2489.
[10]郭峰,曹其新,谢国俊,等. 基于OHTA颜色空间的瓜果轮廓提取方法[J]. 农业机械学报,2005,11:119-122.
[11]吴茜茵,严云洋,杜静,等.多特征融合的火焰检测算法[J]. 智能系统学报,2015,10:240-247.
[12]HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J].The Royal Society, 1998, 454(1971):903-995.
[13]NUNES J C, BOUAOUNE Y, DELECHELLE E, et al. Image analysis by bidimensional empirical mode decomposition[J]. Image and Vision Computing,2003,21(12):1019-1026.
[14]沈滨,崔峰,彭思龙.二维EMD的纹理分析及图像瞬时频率估计[J].计算机辅助设计与图形学学报,2005,17(10): 2345-2351.
[15]OJALA T, PIETIKAINEN M, MAENPAA T. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971-987.
[16]严云洋, 唐岩岩, 刘以安, 等. 使用多尺度LBP 特征和SVM的火焰识别算法[J]. 山东大学学报(工学版), 2012(5): 47-52,58.
[17]HSIEH C J, CHANG K W, LIN C J, et al. A dual coordinate descent method for large-scale linear SVM[J]. Icml, 2015, 9(3):1369-1398.
[18]CHEN T H, WU P H, CHIOU Y C. An early fire-detection method based on image processing[C]∥Proceedings of the 2004 IEEE International Conference on Image Processing(ICIP04). Washington D C: IEEE Computer Society Press, 2004:1707-1710.
[19]RONG J Z, ZHOU D C, YAO W, et al. Fire flame detection based on GICA and target tracking[J]. Optics & Laser Technology, 2013,47:283-291.

备注/Memo

备注/Memo:
收稿日期:2016-09-08。
基金项目:国家自然科学基金资助项目(61402192); 江苏省“六大人才高峰”项目(2013DZXX-023); 江苏省“333工程”(BRA2013208); 淮安市科技计划项目(HAG2013057,HAG2013059)。
作者简介:戴静(1990—), 女,江苏泰州人,硕士生。通讯联系人:严云洋(1967—), E-mail:areyyyke@163.com
更新日期/Last Update: 2017-04-01