[1]严云洋,朱晓妤,刘以安,等.基于Faster R-CNN模型的火焰检测[J].南京师范大学学报(自然科学版),2018,41(03):1.[doi:10.3969/j.issn.1001-4616.2018.03.001]
　Yan Yunyang,Zhu Xiaoyu,Liu Yian,et al.Flame Detection Based on Faster R-CNN Model[J].Journal of Nanjing Normal University(Natural Science Edition),2018,41(03):1.[doi:10.3969/j.issn.1001-4616.2018.03.001]
点击复制

基于Faster R-CNN模型的火焰检测()
分享到：

《南京师范大学学报》（自然科学版）[ISSN:1001-4616/CN:32-1239/N]

卷:
第41卷
期数:
2018年03期
页码:
1
栏目:
·人工智能算法与应用专栏·
出版日期:
2018-09-30

文章信息/Info

Title:
Flame Detection Based on Faster R-CNN Model
文章编号:
1001-4616(2018)03-0001-05
作者:
严云洋1; 2; 朱晓妤1; 2; 刘以安2; 高尚兵1
(1.淮阴工学院计算机与软件工程学院,江苏 淮安 223003)(2.江南大学物联网工程学院,江苏 无锡 214122)
Author(s):
Yan Yunyang1; 2; Zhu Xiaoyu1; 2; Liu Yi’an2; Gao Shangbing1
(1.Faculty of Computer & Software Engineering,Huaiyin Institute of Technology,Huaian 223003,China)(2.School of Internet of Things Engineering,Jiangnan University,Wuxi 214122,China)
关键词:
Faster R-CNN; 候选区域生成网络; 快速区域卷积神经网络; 火焰检测
Keywords:
Faster R-CNN; RPN; Fast R-CNN; flame detection
分类号:
TP391.41
DOI:
10.3969/j.issn.1001-4616.2018.03.001
文献标志码:
A
摘要:
常规的火焰检测一般是提取火焰的静态或动态特征,然后进行火焰的判别. 但是传统特征无法全面描述火焰特性,会导致识别的准确率降低. 本文提出一种基于Faster R-CNN模型的火焰检测算法. 首先利用候选区域生成网络(Region Proposal Network,RPN)提取火焰候选区域,然后对候选区域进行卷积及池化操作,提取火焰特征,最后利用联合训练的快速区域卷积神经网络(Fast R-CNN)进行火焰识别. 实验结果表明该方法能够自动提取火焰特征,有效提高复杂背景下的火焰识别的准确率,具有良好的泛化能力和鲁棒性.
Abstract:
Usually the static or dynamic characteristics of the flame is extracted for flame detection. But the traditional characteristics can not fully describe the characteristics of flame,which leads to the reduction of recognition accuracy. To solve this problem,a flame detection based on Faster R-CNN model is proposed in this paper. First,the candidate region of the flame is extracted by RPN. Then the convolution and pool operation of candidate regions are performed to extract the flame characteristics. Finally,Fast R-CNN is used to identify the flame. The experimental results show that the method can automatically extract the flame characteristics,effectively improve the accuracy of flame recognition in the complex background,and have good generalization ability and robustness.

参考文献/References:

[1] 相徐斌. 基于视频的烟雾检测算法研究[D]. 杭州:浙江大学,2017.
[2]熊国良,苏兆熙,刘举平,等. 火焰特性识别的Matlab实现方法[J]. 计算机工程与科学,2013,35(7):131-136.
[3]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.
[4]胡勤,陈琛,刘敏. 一种基于动态纹理的烟雾和火焰检测方法[J]. 消防科学与技术,2014,33(6):667-669.
[5]陈磊,黄继风. 基于视频的火焰检测方法[J]. 计算机工程与设计,2014,35(9):3143-3147.
[6]STADLER A,WINDISCH T,DIEPOLD K. Comparison of intensity flickering features for video based flame detection algorithms[J]. Fire safety journal,2014,66:1-7.
[7]王华利,邹俊忠,张见,等. 基于深度卷积神经网络的快速图像分类算法[J]. 计算机工程与应用,2017,53(13):181-188.
[8]GIRSHICK R,DONAHUE J,DARRELL T,et al. Region-based convolutional networks for accurate object detection and segmentation[J]. IEEE Trans Pattern Anal Mach Intell,2015,38(1):142-158.
[9]曹诗雨,刘跃虎,李辛昭. 基于Fast R-CNN的车辆目标检测[J]. 中国图象图形学报,2017,22(5):671-677.
[10]REN S,HE K,GIRSHICK R,et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis & machine intelligence,2017,39(6):1137.
[11]JIA Y,SHELHAMER E,DONAHUE S,et al. Caffe:convolutional architecture for fast feature embedding[C]//Proceedings of the 22nd ACM International Conference on Multimedia. Orlando,Florida:ACM,2014.
[12]CHMELAR P,BENKRID A. Efficiency of HSV over RGB Gaussian Mixture Model for fire detection[C]//Radioelektronika. Slovakia:IEEE,2014.
[13]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.
[14]FRIZZI S,KAABI R,BOUCHOUICHA M,et al. Convolutional neural network for video fire and smoke detection[C]//Industrial Electronics Society,IECON 2016-42nd Annual Conference of the IEEE. USA:IEEE,2016.

备注/Memo

备注/Memo:
收稿日期:2018-04-16.
基金项目:国家自然科学基金(61402192)、江苏省“六大人才高峰”项目(2013DZXX-023)、江苏省“青蓝工程”、淮安市“533英才工程”.
通讯联系人:严云洋,博士,教授,研究方向:数字图像处理、模式识别. E-mail:areyyyke@163.com
更新日期/Last Update: 2018-11-19