[1]骈璐璐,裴焕斗,张宇璇.多场景烟雾环境下改进的YOLOv5s烟雾检测算法[J].工业仪表与自动化装置,2024,(02):101-105.[doi:DOI:10.19950/j.cnki.CN61-1121/TH.2024.02.018]
　PIAN Lulu,PEI Huandou,ZHANG Yuxuan.Improved YOLOv5s smoke detection algorithm in multi-scenario smoke environment[J].Industrial Instrumentation & Automation,2024,(02):101-105.[doi:DOI:10.19950/j.cnki.CN61-1121/TH.2024.02.018]

点击复制

多场景烟雾环境下改进的YOLOv5s烟雾检测算法(PDF)

《工业仪表与自动化装置》[ISSN:1000-0682/CN:61-1121/TH]

卷:

期数:

2024年02期

页码:

101-105

栏目:

出版日期:

2024-04-15

文章信息/Info

Title:

Improved YOLOv5s smoke detection algorithm in multi-scenario smoke environment

文章编号:

1000-0682（2024）02-0101-05

作者:

骈璐璐; 裴焕斗; 张宇璇

(1.中北大学 仪器与电子学院，山西 太原 030051; 2.百信信息技术有限公司，山西 太原 030006）

Author(s):

PIAN Lulu1; 2 ; PEI Huandou1; 2 ; ZHANG Yuxuan1; 2

（1. School of Instrument and Electronics, North University of China, shanxi Taiyuan 030051, China; 2. 100 Trust Information Technology Co., LTD., shanxi Taiyuan 030006, China)

关键词:

深度学习; 图像处理; PSA; 烟雾检测

Keywords:

deep learning; image processing; PSA; smoke detection

分类号:

TP391

DOI:

DOI:10.19950/j.cnki.CN61-1121/TH.2024.02.018

文献标志码:

A

摘要:

烟雾检测往往检测精度低、漏检率与误检率高，为了解决这些问题，提出一种改进的YOLOv5s的烟雾检测模型。首先将C3_PSA结构添加到YOLOv5s的骨干部分中，提高模型在恶劣的环境下对烟雾边界的特征提取能力，降低漏检率；其次，将BiFusion的结构替换YOLOv5s模型的颈部结构，增强模型对细节特征以及位置信息的检测能力，提高模型检测精度，降低模型误检率；最后，对损失函数进行改进以进一步提高模型检测准确率。实验结果显示，改进后的模型的精度提升了5.6%，平均精度均值（mAP）提升了3.5%，FPS为369帧/秒，说明改进后的模型可以在背景环境复杂时准确地检测出烟雾特征，同时满足模型在端侧部署的实时性高的要求。

Abstract:

Smoke detection often has low detection accuracy, high missed detection rate and false detection rate, in order to solve these problems, an improved YOLOv5s smoke detection model was proposed. Firstly, the C3_PSA structure was added to the backbone of YOLOv5s to improve the feature extraction ability of the model in harsh environments and reduce the missed detection rate. Secondly, BiFusion structure was used to replace the neck structure of YOLOv5s model, so as to enhance the model’s ability to detect details and location information, improve the model detection accuracy, and reduce the model false detection rate. Finally, the loss function is improved to further improve the accuracy of model detection. The experimental results show that the accuracy of the improved model is improved by 5.6%, the mAP is improved by 3.5%, and the FPS is 369 frames/second, which indicates that the improved model can accurately detect the smoke characteristics in the complex background environment, and at the same time meet the requirements of high real-time deployment of the model on the end side.

参考文献/References:

[1]王韦刚,王炳蔚,张云伟.TDFF:一种强鲁棒性的烟雾图像检测算法[J].激光与光电子学进展,2021,58(04):273-280.

相似文献/References:

[1]薛 彪,张松飞,赵 莉.基于单目视觉的近景摄影测量在振动台试验中的应用[J].工业仪表与自动化装置,2015,(03):44.
　XUE Biao,ZHANG Songfei,ZHAO Li.Application of close-range photogrammetry based on monocular vision in shaking table test[J].Industrial Instrumentation & Automation,2015,(02):44.
[2]辛 迈,王 平,曾瑞鹏,等.基于线性插值的图像分块分割算法[J].工业仪表与自动化装置,2016,(01):118.
　XIN Mai,WANG Ping,ZENG Ruipeng,et al.Image segmentation algorithms based on linear interpolation of blocks[J].Industrial Instrumentation & Automation,2016,(02):118.
[3]杨 帅,张有芬,李玉惠,等.基于深度卷积神经网络的车标分类[J].工业仪表与自动化装置,2017,(05):75.
　YANG Shuai,ZHANG Youfen,LI Yuhui,et al.Vehicle classification based on deep convolutional neural network[J].Industrial Instrumentation & Automation,2017,(02):75.
[4]张永刚,陈 军*.基于模型法的医学影像图像纹理分析研究[J].工业仪表与自动化装置,2018,(03):101.[doi:1000-0682（2018）03-0000-00]
　ZHANG Yonggang,CHEN Jun*.Analysis of medical image texture based on the model method[J].Industrial Instrumentation & Automation,2018,(02):101.[doi:1000-0682（2018）03-0000-00]
[5]徐先峰,黄刘洋,龚 美.基于卷积神经网络与双向长短时记忆网络组合模型的短时交通流预测[J].工业仪表与自动化装置,2020,(01):13.
　XU Xianfeng,HUANG Liuyang,GONG Mei.Short-term traffic flow prediction based on combined model of convolutional neural network and bidirectional long-term memory network[J].Industrial Instrumentation & Automation,2020,(02):13.
[6]张晓华,马 煜,杨晨辉,等.基于卷积神经网络的设备安装位置智能识别方法[J].工业仪表与自动化装置,2021,(01):13.[doi:10.3969/j.issn.1000-0682.2021.01.003]
　ZHANG Xiaohua,MA yu,YANG Chenhui,et al.Intelligent identification method of equipment installation position based on convolution neural network[J].Industrial Instrumentation & Automation,2021,(02):13.[doi:10.3969/j.issn.1000-0682.2021.01.003]
[7]苏 杨,卢 翔,李 琨,等.基于轻量深度学习网络的机房人物检测研究[J].工业仪表与自动化装置,2021,(01):100.[doi:10.3969/j.issn.1000-0682.2021.01.024]
　SU Yang,LU Xiang,LI Kun,et al.Research on computer room human detection based on lightweight deep learning network[J].Industrial Instrumentation & Automation,2021,(02):100.[doi:10.3969/j.issn.1000-0682.2021.01.024]
[8]李瑞丹,王国庆,苑文峰,等.紧凑型桌面贴片机系统研究[J].工业仪表与自动化装置,2021,(02):81.[doi:10.19950/j.cnki.cn61-1121/th.2021.02.018]
　LI Ruidan,WANG Guoqing,YUAN Wenfeng,et al.Research on compact desktop mounter system[J].Industrial Instrumentation & Automation,2021,(02):81.[doi:10.19950/j.cnki.cn61-1121/th.2021.02.018]
[9]卢 翔,苏 杨,余 萱,等.基于深度学习的机房人物重识别研究[J].工业仪表与自动化装置,2021,(02):104.[doi:10.19950/j.cnki.cn61-1121/th.2021.02.024]
　LU Xiang,SU Yang,YU Xuan,et al.Research on computer room character recognition based on deep learning[J].Industrial Instrumentation & Automation,2021,(02):104.[doi:10.19950/j.cnki.cn61-1121/th.2021.02.024]
[10]倪四清,左光恒,张 俊.高速公路建设远程视频监控系统的设计与实现[J].工业仪表与自动化装置,2022,(02):76.[doi:10.19950/j.cnki.cn61-1121/th.2022.02.016]
　NI Siqing,ZUO Guangheng,ZHANG Jun.Design and realization of remote video monitoring system for expressway construction[J].Industrial Instrumentation & Automation,2022,(02):76.[doi:10.19950/j.cnki.cn61-1121/th.2022.02.016]

备注/Memo

备注/Memo:

收稿日期：2023-12-06第一作者：骈璐璐（1999—），女，河南安阳人，硕士，研究方向为计算机视觉。

常用功能