[1]严 忱,严云洋,高尚兵,等.基于多级特征融合的视频火焰检测方法[J].南京师大学报(自然科学版),2021,44(03):131-136.[doi:10.3969/j.issn.1001-4616.2021.03.019]
　Yan Chen,Yan Yunyang,Gao Shangbing,et al.Video Flame Detection Based on Fusion of Multilevel Features[J].Journal of Nanjing Normal University(Natural Science Edition),2021,44(03):131-136.[doi:10.3969/j.issn.1001-4616.2021.03.019]
点击复制

基于多级特征融合的视频火焰检测方法()
分享到：

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

卷:
第44卷
期数:
2021年03期
页码:
131-136
栏目:
·计算机科学与技术·
出版日期:
2021-09-15

文章信息/Info

Title:
Video Flame Detection Based on Fusion of Multilevel Features
文章编号:
1001-4616(2021)03-0131-06
作者:
严 忱1; 严云洋1; 2; 高尚兵1; 朱全银1
(1.淮阴工学院计算机与软件工程学院,江苏 淮安 223003)(2.江苏海洋大学计算机工程学院,江苏 连云港 225005)
Author(s):
Yan Chen1; Yan Yunyang1; 2; Gao Shangbing1; Zhu Quanyin1
(1.Faculty of Computer & Software Engineering,Huaiyin Institute of Technology,Huai’an 223003,China)(2.School of Computer Engineering,Jiangsu Ocean University,Lianyungang 225005,China)
关键词:
火焰检测; 目标检测; 特征融合; Yolov2
Keywords:
flame detection; object detection; feature fusion; Yolov2
分类号:
TP391
DOI:
10.3969/j.issn.1001-4616.2021.03.019
文献标志码:
A
摘要:
火焰前期一般多为小目标,但一般的火焰检测方法对于小目标的检测能力较差. 为检测早期火焰,提高火灾预防能力,提出了一种融合多级特征的视频火焰检测方法,针对下采样分辨率变小导致丢失目标的问题,引入了反卷积模块,并融合深层具有较强语义信息的特征和浅层具有较强细节信息的特征,从而有效提高了火焰的检测率. 所提算法在Bilkent大学火灾数据库VisiFire数据集上进行的实验表明,mAP相较于Yolov2提高了10.0%,与多种经典的深度学习算法模型相比,检测率更高.
Abstract:
In the early stage,the flame is usually small target,but the general flame detection method has poor detection ability for small target. To detect early flame and improve fire prevention ability,a video flame detection method with multi-stage features is proposed. Aiming at the problem of missing target caused by the lower sampling resolution,a deconvolution module is introduced,and the features with strong semantic information in the deep layer and strong detail information in the shallow layer are combined to effectively improve the detection rate of flame. The experiment of the proposed algorithm on the Fire database VisiFire data set of Bilkent University showed that mAP improved by 10.0% compared with the Yolov2,and the detection rate was higher compared with many classic deep learning algorithm models.

参考文献/References:

[1] 洪柳. 建筑火灾事故损失特性研究[J]. 建筑安全,2020,35(1):59-64.
[2]HOWARD A G,ZHU M,CHEN B,et al. Mobilenets:efficient convolutional neural networks for mobile vision applications[J]. arXiv preprint arXiv:1704.04861,2017.https://arxiv.org/abs/1704.04861.
[3]SHARMA J,GRANMO O C,GOODWIN M. Deep CNN-ELM hybrid models for fire detection in images[C]//International Conference on Artificial Neural Networks. Cham:Springer,2018:245-259.
[4]LEE W,KIM S,LEE Y T,et al. Deep neural networks for wild fire detection with unmanned aerialehicle[C]//IEEE International Conference on Consumer Electronics. Las Vegas,USA:IEEE,2017.
[5]ZHANG D,HAN S,ZHAO J,et al. Image based forest fire detection using dynamic characteristics with artificial neural networks[C]//Iita International Joint Conference on Artificial Intelligence. Hainan Island,China,2009:290-293.
[6]黄文锋,徐珊珊,孙燚等. 基于多分辨率卷积神经网络的火焰检测[J]. 郑州大学学报(工学版),2019,40(5):80-84.
[7]江洋,白勇. 基于RetinaNet深度学习模型的火焰检测研究[J/OL]. 海南大学学报(自然科学版):1-7[2019-12-12]. http://kns.cnki.net/kcms/detail/46.1013.N.20191119.1528.010.html.
[8]LIU W,ANGUELOV D,ERHAN D,et al. SSD:single shot multibox detector[C]//European Conference on Computer Vision. Cham:Springer,2016:21-37.
[9]REDMON J,DIVVALA S,GIRSHICK R,et al. You only look once:unified,real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas,USA:IEEE,2016:779-788.
[10]FU C Y,LIU W,RANGA A,et al. Dssd:deconvolutional single shot detector[J]. arXiv preprint arXiv:1701.06659,2017.
[11]REDMON J,FARHADI A. Yolov3:an incremental improvement[J]. arXiv preprint arXiv:1804.02767,2018.
[12]BOCHKOVSKIY A,WANG C Y,LIAO H Y M. YOLOv4:optimal speed and accuracy of object detection[J]. Computer vision and pattern recognition,2020,17(9):198-215.
[13]REDMON J,FARHADi A. YOLO9000:better,faster,stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu,USA:IEEE,2017:7263-7271.
[14]张旭,李建胜,郝向阳,等. 基于差分筛选的YOLOv2监控视频目标检测方法[J]. 测绘科学技术学报,2018,35(6):616-621.
[15]REN S,HE K,GIRSHICK R,et al. Faster R-CNN:towards real-time object detection with region proposal networks[C]//Advances in Neural Information Processing Systems. Montreal,2015:91-99.
[16]黄同愿,向国徽,杨雪姣. 基于深度学习的行人检测技术研究进展[J]. 重庆理工大学学报(自然科学),2019,33(4):98-109.

相似文献/References:

[1]严云洋,朱晓妤,刘以安,等.基于Faster R-CNN模型的火焰检测[J].南京师大学报(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]
[2]郑 冬,李向群,许新征.基于轻量化SSD的车辆及行人检测网络[J].南京师大学报(自然科学版),2019,42(01):73.[doi:10.3969/j.issn.1001-4616.2019.01.012]
　Zheng Dong,Li Xiangqun,Xu Xinzheng.Vehicle and Pedestrian Detection Model Based on Lightweight SSD[J].Journal of Nanjing Normal University(Natural Science Edition),2019,42(03):73.[doi:10.3969/j.issn.1001-4616.2019.01.012]
[3]蔡钟晟,陈 飞,曾勋勋.一种具有抗噪性能的圆形目标检测器[J].南京师大学报(自然科学版),2021,44(04):85.[doi:10.3969/j.issn.1001-4616.2021.04.011]
　Cai Zhongsheng,Chen Fei,Zeng Xunxun.A Circular Object Detector with Anti-Noise Performance[J].Journal of Nanjing Normal University(Natural Science Edition),2021,44(03):85.[doi:10.3969/j.issn.1001-4616.2021.04.011]

备注/Memo

备注/Memo:
收稿日期:2021-05-20.
基金项目:国家自然科学基金项目(61402192)、江苏省“六大人才高峰”项目(2013DZXX-023)、江苏省高校自然科学基金重大项目(18KJA52001).
通讯作者:严云洋,博士,教授,研究方向:数字图像处理、模式识别. E-mail:yuhyang@hyit.edu.cn
更新日期/Last Update: 2021-09-15