[1]乐其河,陈炜,郑祥盘,等.改进YOLOv8n模型的火灾场景火焰检测方法[J].华侨大学学报(自然科学版),2025,46(3):255-263.[doi:10.11830/ISSN.1000-5013.202412006]
　LE Qihe,CHEN Wei,ZHENG Xiangpan,et al.Flame Detection Method in Fire Scene With Improved YOLOv8n Model[J].Journal of Huaqiao University(Natural Science),2025,46(3):255-263.[doi:10.11830/ISSN.1000-5013.202412006]
点击复制

改进YOLOv8n模型的火灾场景火焰检测方法()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第46卷
期数:
2025年第3期
页码:
255-263
栏目:
出版日期:
2025-05-20

文章信息/Info

Title:
Flame Detection Method in Fire Scene With Improved YOLOv8n Model
文章编号:
1000-5013(2025)03-0255-09
作者:
乐其河1; 2 ;  陈炜1;  郑祥盘1;  许亦镜1;  林立霖1
1. 闽江学院 物理与电子信息工程学院, 福建 福州 350108;2. 华侨大学 机电及自动化学院, 福建 厦门 361021
Author(s):
LE Qihe1; 2 ;  CHEN Wei1;  ZHENG Xiangpan1; XU Yijing1;  LIN Lilin1
1. College of Physics and Electronic Information Engineering, Minjiang University, Fuzhou 350108, China; 2. College of Mechanical Engineering and Automation, Huaqiao University, Xiamen 361021, China
关键词:
火焰检测;  椒盐噪声;  YOLOv8n模型;  中值滤波模块;  轻量级Ghost卷积
Keywords:
flame detection;  random noise;  YOLOv8n model;  median filtering module;  lightweight Ghost convolution
分类号:
TP391
DOI:
10.11830/ISSN.1000-5013.202412006
文献标志码:
A
摘要:
针对火灾复杂烟尘环境导致火焰检测准确性低下的问题,提出一种基于YOLOv8n模型的高效精准火焰检测方法。首先,选取多种火灾场景图像作为数据集原始图像,同时增加随机椒盐等噪声模拟烟尘环境;其次,在模型的网络框架前端嵌入中值滤波模块,旨在提升烟尘环境下网络对干扰噪声的理解能力;最后,利用Ghost卷积模块,设计不同层级的跨层连接网络,在减少参数量的同时,优化了网络的泛化能力,实现了在噪声干扰的火灾场景下实时高精的火焰检测。实验结果表明:改进的YOLOv8n模型具有更优异的实时性和检测准确性。
Abstract:
Aiming at the problem of low accuracy in flame detection caused by complex smoke and dust environments in fire scenes, an efficient and precise flame detection method based on the YOLOv8n model was proposed. First, a variety of fire scene images were selected as the original images for the dataset, and random noise, such as salt and pepper noise, was added to simulate a smoke and dust environment. Second, a median filtering module was embedded at the front of the model’s network framework to enhance the network’s capability to handle interference noise in smoke and dust environments. Finally, by utilizing Ghost convolution modules and designing cross layer connection networks at different lay levels,the number of parameters was reduced while the generalization capability of the network was optimized. This enable real-time and high-precision flame detection in fire scene with noise interference. Experimental results show that the improved YOLOv8n model had superior real-time performance and detection accuracy performance.

参考文献/References:

[1] 卞苏阳,严云洋,龚成张,等.基于CXANet-YOLO的火焰检测方法[J].南京大学学报(自然科学版),2023,59(2):295-301.DOI:10.13232/j.cnki.jnju.2023.02.012.
[2] 周兴华,陈西江,羊海东,等.动态卷积YOLOv5的视频火焰检测算法[J].测绘科学,2023,48(4):106-118.DOI:10.16251/j.cnki.1009-2307.2023.04.012.
[3] 李伟达,叶靓玲,郑力新,等.面向扶梯不安全行为的改进型深度学习检测算法[J].华侨大学学报(自然科学版),2022,43(1):119-126.DOI:10.11830/ISSN.1000-5013.202105059.
[4] SATHISHKUMAR V E,CHO J,SUBRAMANIAN M,et al.Forest fire and smoke detection using deep learning-based learning without forgetting[J].Fire Ecology,2023,19(1):9.DOI:10.1186/s42408-022-00165-0.
[5] 崔志亮,曹苏群.基于改进YOLOv5的复杂环境下火灾检测方法[J].电脑与信息技术,2024,32(1):23-27,46.DOI:10.19414/j.cnki.1005-1228.2024.01.019.
[6] 杜建华,张认成.LVQ神经网络的红外光谱火灾早期预警算法[J].华侨大学学报(自然科学版),2011,32(6):607-610.DOI:10.11830/ISSN.1000-5013.2011.06.0607.
[7] CHENG Guangtao,CHEN Xue,WANG Chenyi,et al.Visual fire detection using deep learning: A survey[J].Neurocomputing,2024,596:127975.DOI:10.1016/j.neucom.2024.127975.
[8] 蔡春兵,吴翠平,徐鲲鹏.基于深度学习的视频火焰识别方法[J].信息技术与网络安全,2020,39(12):44-51.DOI:10.19358/j.issn.2096-5133.2020.12.008.
[9] 王勇智,谭杨磊,韩锐,等.基于分离YUV颜色通道的火焰提取方法[J].科学技术创新,2020(25):42-43.
[10] 何建,黄亦豪,薛黎明.基于YCbCr和帧间差分法的火焰分割算法[J].现代计算机,2023,29(18):49-52.
[11] ZHANG Heng,SHAO Faming,CHU Weijun,et al.Faster R-CNN based on frame difference and spatiotemporal context for vehicle detection[J].Signal Image and Video Processing,2024,18:7013-7027.DOI:10.1007/s11760-024-03370-3.
[12] YADAV A,KUMAR E.Objectdetection on real-time videowith fpn and modified mask rcnnbased on inception-resnet V2[J].Wireless Personal Communications,2024,138(4):2065-2090.DOI:10.1007/s11277-024-11539-9.
[13] LIU Wei,ANGUELOV D,ERHAN D,et al.SSD: Single shot multibox detector[M].Cham:Springer International Publishing,2016:21-37.DOI:10.1007/978-3-319-46448-0_2.
[14] REDMON J,DIVVALA S,GIRSHICK R,et al.You onlylook once: Unified, real-time object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Press,2016:779-788.DOI:10.48550/arXiv.1506.02640.
[15] REDMON J,FARHADI A.YOLOv3: An incremental improvement[EB/OL].(2018-04-08)[2024-02-21] .https://arxiv.org/pdf/1804.02767.
[16] BOCHKOVSKIY A,WANG C Y,LIAO H M.YOLOv4: Optimalspeed and accuracy of object detection[EB/OL].(2020-04-23)[2024-02-21] .https://www.researchgate.net/publication/340883401_YOLOv4_Optimal_Speed_and_Accuracy_of_Object_Detection.
[17] QIN Yueyan,CAO Jiangtao,JI Xiaofei.Fire detection method based on depthwise separable convolution and YOLOv3[J].International Journal of Automation and Computing,2021,18(2):300-310.DOI:10.1007/s11633-020-1269-5.
[18] 汪子健,高焕兵,侯宇翔,等.改进YOLOX-nano的火灾火焰烟雾检测[J].计算机系统应用,2023,32(3):265-274.DOI:10.15888/j.cnki.csa.009000.
[19] 曹康壮,焦双健.融合注意力机制的轻量级火灾检测模型[J].消防科学与技术,2024,43(3):378-383.
[20] 陈义啸,沈景凤,仲梁维.基于改进YOLOv8的森林火灾检测算法研究[J].软件工程,2024,27(2):49-54.DOI:10.19644/j.cnki.issn2096-1472.2024.002.010.
[21] 王菲.基于火焰识别的早期火灾探测技术研究[D].广州:华南理工大学,2013.
[22] 陈智铭,谢维波,许华滨.利用图像频度特征的稀薄烟雾检测算法[J].华侨大学学报(自然科学版),2014,35(2):151-155.DOI:10.11830/ISSN.1000-5013.2014.02.0151.
[23] Lü Changzhi,ZHOU Haiyong,CHEN Yu,et al.A lightweight fire detection algorithm for small targets based on YOLOv5s[J].Scientific Reports,2024,14(1):14104.DOI:10.1038/s41598-024-64934-4.
[24] TALAAT F M,ZAINELDIN H.An improved fire detection approach based on YOLO-v8 for smart cities[J].Neural Computingand Applications,2023,35(28):20939-20954.DOI:10.1007/s00521-023-08809-1.
[25] WANG C Y,BOCHKOVSKIY A,LIAO H Y M.YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//lEEE Conference on Computer Vision and Pattern Recognition.Vancouver:IEEE Press,2022:7464-7475.DOI:10.48550/arXiv.2207.02696.
[26] HAN Kai,WANG Yunhe,TIAN Qi,et al.Ghostnet: More features from cheapoperations[C]//IEEE/CVF Conference on Computer Visionand Pattern Recognition.New York:IEEE Press,2020:1580-1589.DOI:10.48550/arXiv.1911.11907.
[27] LI Jichao,GUO Shengyu,KONG Liulin,et al.An improved YOLOv3-tiny method for fire detection in the construction industry[C]//E3S Web of Conference.[S.l.]:EEM,2021:03069.DOI:10.1051/e3sconf/202125303069.
[28] 谌颃,张袖斌,肖斌,等.基于Faster-R CNN深度学习算法的图像识别技术研究[J].机器人产业,2024(3):98-102.DOI:10.19609/j.cnki.cn10-1324/tp.2024.03.019.
[29] 郑文秀,赵兴娜.基于SSD算法的人脸检测算法研究[J].现代信息科技,2024,8(19):17-22.DOI:10.19850/j.cnki.2096-4706.2024.19.004.
[30] ZHANG Long,LI Jiaming,ZHANG Fuquan.An efficient forest fire target detection model based on improved YOLOv5[J].Fire-Switzerland,2023,6(8):291.DOI:10.3390/fire6080291.

相似文献/References:

[1]陈华玲,冯桂.数字图像的混合噪声去除[J].华侨大学学报(自然科学版),2011,32(2):150.[doi:10.11830/ISSN.1000-5013.2011.02.0150]
　CHEN Hua-ling,FENG Gui.A Method of Mixed Noise Removal in Digital Image[J].Journal of Huaqiao University(Natural Science),2011,32(3):150.[doi:10.11830/ISSN.1000-5013.2011.02.0150]

备注/Memo

备注/Memo:
收稿日期: 2024-12-06
通信作者: 陈炜(1991-),男,讲师,博士,主要从事机器视觉算法、深度学习方法的研究。E-mail:chenwei.edu@outlook.com。
基金项目: 福建省自然科学基金资助项目(2022J05235); 福建省技术创新重点攻关及产业化项目(校企联合类)(2023XQ018); 闽江学院“揭榜挂帅”项目(ZD202303); 闽江学院预研项目(MJY22022)http://hdxb.hqu.edu.cn/
更新日期/Last Update: 2025-05-20