改进YOLOv5的隧道火灾帧差检测网络与应用方法

doi:10.3778/j.issn.1002-8331.2205-0451

摘要/Abstract

摘要： 隧道发生火灾存在着检测难、救援难的问题，实时的火灾监测对于及时发现火情是至关重要的。传统基于视频图像的火灾检测方法，检测依赖单幅图像，无法提取多幅图像的时空信息，检测精度低，不能有效检测隧道火灾。因此，提出了隧道火灾帧差网络。帧差网络使用3D卷积核构建网络结构，提取视频中火灾的时间上下文信息；将帧差网络衔接至YOLOv5主干网络形成隧道火灾帧差检测网络，可以检测单幅图像及两幅图像，从而充分利用视频动态信息；使用CIoU函数优化网络的边界框损失，并融合分类损失与置信度损失，使网络能够快速收敛。实验结果表明，该网络在隧道火灾数据集上的平均精度高达91.03%，检测速度达到了63.7帧/s，具有较强的鲁棒性。通过选取最优分析策略设计隧道火灾检测应用方法，该方法在隧道场景中的漏检率和误检率分别为2.52%和2.03%，可以满足隧道火灾检测的准确性和实时性需求。

关键词: 隧道, 火灾检测, 帧差网络, YOLOv5

Abstract: Fires in tunnels have the problem of difficult detection and rescue. Thus real-time fire monitoring is crucial for prompt discovery of fires. Traditional video image-based fire detection methods rely on a single image and cannot extract temporal and spatial information from multiple images. The detection precision is limited and not effective in detecting tunnel fires. As a result, a tunnel fire frame difference network is proposed. The structure of the frame difference network is mainly a 3D convolutional kernel. Temporal contextual information about the fire in the video can be extracted. The network connects to the YOLOv5 backbone network to form tunnel fire detection network, which can detect a single image as well as two images. This takes full advantage of the motion information of the videos. The network achieves fast convergence by optimizing the bounding box loss using the CIoU function and combining classification loss and confidence loss. Experimental results show that the network achieves an average precision of 91.03% on the tunnel fire dataset, with a detection speed of 63.7 frames per second. The model has strong robustness. The network and analysis strategies are combined to form a tunnel fire detection system. It has a false detection rate and a missed detection rate of 2.52% and 2.03% in the tunnel scene, respectively. The results can satisfy the accurate and real-time requirements of tunnel fire detection.

Key words: tunnel, fire detection, frame difference network, YOLOv5

张晋瑞, 宋焕生, 孙士杰, 梁浩翔, 张朝阳, 王宇, 刘莅辰. 改进YOLOv5的隧道火灾帧差检测网络与应用方法[J]. 计算机工程与应用, 2023, 59(2): 222-231.

ZHANG Jinrui, SONG Huansheng, SUN Shijie, LIANG Haoxiang, ZHANG Zhaoyang, WANG Yu, LIU Lichen. Tunnel Fire Frame Difference Detection Network by Improving YOLOv5 and Application Method[J]. Computer Engineering and Applications, 2023, 59(2): 222-231.

参考文献

[1] 2020年交通运输行业发展统计公报[J].交通财会，2021（6）：92-97.
2020 statistical bulletin on the development of the Transportation industry[J].Finance & Accounting for Transport，2021（6）：92-97.
[2] LEE Y，SHIM J.Deep learning and color histogram based fire and smoke detection research[J].International Journal of Advanced Smart Convergence，2019，8（2）：116-125.
[3] 刘恺，刘湘，常丽萍，等.基于YUV颜色空间和多特征融合的视频烟雾检测[J].传感技术学报，2019（2）：237-243.
LIU K，LIU X，CHANG L P，et al.Video smoke detection based on YUV color space and multiple feature fusion[J].Chinese Journal of Sensors and Actuators，2019（2）：237-243.
[4] 殷梦霞，王理，孙连营.基于多特征融合的自适应烟雾检测算法[J].建筑科学，2019（9）：26-31.
YIN M X，WANG L，SUN L Y.Adaptive smoke detection algorithm based on multi-feature fusion[J].Building Science，2019（9）：26-31.
[5] XIONG D，YAN L.Early smoke detection of forest fires based on SVM image segmentation[J].Journal of Forest Science，2019，65（4）：150-159.
[6] PARK K M，BAE C O.Smoke detection in ship engine rooms based on video images[J].IET Image Processing，2020，14（6）：1141-1149.
[7] FRIZZI S，KAABI R，BOUCHOUICHA M，et al.Convolutional neural network for video fire and smoke detection[C]//IECON 2016-42nd Annual Conference of the IEEE Industrial Electronics Society，2016：877-882.
[8] 罗胜，张翔，胡杰，等.时间压缩轨迹特征识别的火灾烟雾检测[J].中国图象图形学报，2019，24（10）：1648-1657.
LUO S，ZHANG X，HU J，et al.Smoke detection by trajectories in condensed images for early fire warning[J].Journal of Image and Graphics，2019，24（10）：1648-1657.
[9] 李鹏，张炎.基于高斯混合模型和卷积神经网络的视频烟雾检测[J].激光与光电子学进展，2019，56（21）：140-146.
LI P，ZHANG Y.Video smoke detection based on Gaussian mixture model and convolutional neural network[J].Laser & Optoelectronics Progress，2019，56（21）：140-146.
[10] 冯路佳，王慧琴，王可，等.基于目标区域的卷积神经网络火灾烟雾识别[J].激光与光电子学进展，2020，57（16）：83-91.
FENG L J，WANG H Q，WANG K，et al.Convolutional neural network fire smoke detection based on target region[J].Laser & Optoelectronics Progress，2020，57（16）：83-91.
[11] ZHANG Q，LIN G，ZHANG Y，et al.Wildland forest fire smoke detection based on faster R-CNN using synthetic smoke images[J].Procedia Engineering，2018，211：441-446.
[12] 石磊，张海刚，杨金锋.基于改进型SSD的视频烟火检测算法[J].计算机应用与软件，2021，38（12）：161-167.
SHI L，ZHANG H G，YANG J F.Video-based fire and smoke detection based on improved SSD[J].Computer Applications and Software，2021，38（12）：161-167.
[13] 罗小权，潘善亮.改进YOLOV3的火灾检测方法[J].计算机工程与应用，2020，56（17）：187-196.
LUO X Q，PAN S L.Improved YOLOV3 fire detection method[J].Computer Engineering and Applications，2020，56（17）：187-196.
[14] NAMOZOV A，CHO Y I.An efficient deep learning algorithm for fire and smoke detection with limited data[J].Advances in Electrical and Computer Engineering，2018，18（4）：121-128.
[15] 谢书翰，张文柱，程鹏，等.嵌入通道注意力的YOLOv4火灾烟雾检测模型[J].液晶与显示，2021，36（10）：1445-1453.
XIE S H，ZHANG W Z，CHENG P，et al.Fire smoke detection model based on YOLOv4 with channel attention[J].Chinese Journal of Liquid Crystals and Displays，2021，36（10）：1445-1453.
[16] DING Z，ZHAO Y，LI A，et al.Spatial-temporal attention two-stream convolution neural network for smoke region detection[J].Fire，2021，4（4）：66.
[17] 赖金星，周慧，程飞，等.公路隧道火灾事故统计分析及防灾减灾对策[J].隧道建设，2017，37（4）：409-415.
LAI J X，ZHOU H，CHENG F，et al.Statistical analysis of fire accidents in highway tunnels and countermeasures for disaster prevention and reduction[J].Tunnel Construction，2017，37（4）：409-415.
[18] WU X，PARK Y，LI A，et al.Smart detection of fire source in tunnel based on the numerical database and artificial intelligence[J].Fire Technology，2021，57（2）：657-682.
[19] AISHWARYA K，PAVITHRA M.Tunnel fire detection using FBG[J].International Journal of Engineering Research Technology，2020，9：3.
[20] YAN B，LI J，ZHANG M，et al.Raman distributed temperature sensor with optical dynamic difference compensation and visual localization technology for tunnel fire detection[J].Sensors，2019，19（10）：2320.
[21] DAI W，JIANG J，DING G，et al.Development and application of fire video image detection technology in China’s road tunnels[J].Civil Engineering Journal，2019，5（1）：1-17.
[22] SARVARI A，MAZINANI S M.A new tunnel fire detection and suppression system based on camera image processing and water mist jet fans[J].Heliyon，2019，5（6）：e01879.
[23] 王琳，姚新，雷丹.公路隧道火灾初期视频火焰检测[J].中国公路学报，2018，31（11）：121-129.
WANG L，YAO X，LEI D.Video-based recognition of earlyfire flame in road tunnel[J].China Journal of Highway and Transport，2018，31（11）：121-129.
[24] 邓实强，丁浩，杨孟，等.基于视频图像的公路隧道火灾烟雾检测[J].隧道建设，2022，42（2）：291-302.
DENG S Q，DING H，YANG M，et al.Fire smoke detection in highway tunnel based on video image[J].Tunnel Construction，2022，42（2）：291-302.
[25] WU X，ZHANG X，HUANG X，et al.A real-time forecast of tunnel fire based on numerical database and artificial intelligence[J].Building Simulation，2022，15（4）：511-524.
[26] LEI X.Object detection for perceptually-degraded environments[D].Pomona：California State Polytechnic University，2020.
[27] ZHANG H，PENG Y，LIU X，et al.Application of object detection on power station environmental anomaly monitoring[C]//2021 IEEE International Conference on Power，Intelligent Computing and Systems（ICPICS），2021：426-430.
[28] WU Z，XUE R，LI H.Real-time video fire detection via modified YOLOv5 network model[J].Fire Technology，2022：1-27.
[29] HUANG P，CHEN M，CHEN K，et al.A combined real-time intelligent fire detection and forecasting approach through cameras based on computer vision method[J].Process Safety and Environmental Protection，2022，164：629-638.
[30] BOCHKOVSKIY A，WANG C Y，LIAO H Y M.Yolov4：optimal speed and accuracy of object detection[J].arXiv：2004.10934，2020.
[31] LIN T Y，DOLLAR P，GIRSHICK R，et al.Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2017：2117-2125.
[32] LIU S，QI L，QIN H，et al.Path aggregation network for instance segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：8759-8768.
[33] ZHENG Z，WANG P，LIU W，et al.Distance-IoU loss：faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2020，34（7）：12993-13000.

[34] ZHU X，SU W，LU L，et al.Deformable DETR：deformable transformers for end-to-end object detection[J].arXiv：2010.04159，2020.

[35] WANG G，HAN S，DING E，et al.Student-teacher feature pyramid matching for unsupervised anomaly detection[J].arXiv：2103.04257，2021.