Article Details

Title · Authors · Abstract · Keywords · DOI · References

Application and Improvement of YOLOv8 in Flame and Smoke Detection

HAN Weijuan1, PING Yi2, YANG Aolin3, YUAN Jinghui2, CUI Erqiang2, DONG Xinjie2

(1.School of Electro-mechanical Engineering, Zhongyuan Institute of Science and Technology, Zhengzhou 450000, China; 2.Department of Information and Communication, Department of Public Security of Henan Province, Zhengzhou 450003, China; 3.School of Public Administration, Henan University of Economics and Law, Zhengzhou 450046, China)

Section：Computer Technology · Vol. 2025, Issue 2 · Pages：1-6 · 2096-4706（2025）02-0001-06 · CLC number：TP391.4 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2025.02.001

Abstract

The fire incidents seriously threaten the safety of people's lives and property, so the fire detection is extremely essential. Based on the YOLOv8 algorithm, this paper carries out the smoke and flame detection, and improves the model structure to raise accuracy. The improvements include 3 aspects of introducing the DBB module, using the Dynamic Convolution, and optimizing the loss function. The experiments demonstrate that the three kinds of improved algorithms all have a specific rise in detection accuracy, and the YOLOv8n model which uses three improvements at the same time has increased mAP50 by 3.03% and mAP50-95 by 3.37% compared to the original YOLOv8n. Compared to Faster R-CNN and other models, various performance indicators such as mAP50 and mAP50-95 have notable enhancements, and the precision of fire detection is also improved.

Keywords

YOLOv8; network structure; convolution; loss function; mAP; accuracy

References

  1. [1] REN S Q，HE K M，GIRSHICK R，et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39（16）：1137-1149.
  2. [2] LIU W，ANGUELOV D，ERHAN D，et al. SSD: Single Shot Multibox Detector [C]//Computer Vision-ECCV 2016.Amsterdam：Springer International Publishing，2016：21-37.
  3. [3] JIANG P Y，ERGU D，LIU F Y，et al. A Review of YOLO Algorithm Developments [J].Procedia Computer Science，2022，199：1066-1073.
  4. [4] 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. 2016：779-788.
  5. [5] Xie Kangkang, Zhu Wenzhong, Xie Linsen, etc Flame and Smoke Detection Algorithm Based on Improved YOLOv7 [J]. Foreign Electronic Measurement Technology, 2023, 42 (7): 41-49
  6. [6] Wang Yixu, Xiao Xiaoling, Wang Pengfei, etc Improved YOLOv5s small target smoke and flame detection algorithm [J]. Computer Engineering and Applications, 2023, 59 (1): 72-81
  7. [7] Song Huawei, Qu Xiaojuan, Yang Xin, etc Flame and smoke detection based on improved YOLOv5 [J]. Computer Engineering, 2023, 49 (6): 250-256
  8. [8] ZHOU Y J，ZHU W G，HE Y H，et al. Yolov8-based Spatial Target Part Recognition [C]//2023 IEEE 3rd International Conference on Information Technology，Big Data and Artificial Intelligence(ICIBA).Chongqing：IEEE，2023：1684-1687.
  9. [9] Yuan Hongchun, Tao Lei Detection and Identification of Fish in Improved Yolov8 Commercial Fishing Vessel Electronic Monitoring Data [J]. Journal of Dalian Ocean University, 2023, 38 (3): 533-542
  10. [10] DING X H，ZHANG X Y，HAN J G，et al. Diverse Branch Block: Building a Convolution as an Inception-like Unit [C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Nashville：IEEE，2021：10886-10895.
  11. [11] SZEGEDY C，VANHOUCKE V，IOFFE S，et al. Rethinking the Inception Architecture for Computer Vision [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas：IEEE，2016：2818-2826.
  12. [12] DAI J F，QI H Z，XIONG Y W，et al. Deformable Convolutional Networks [C]//Proceedings of the IEEE International Conference on Computer Vision.Venice：IEEE，2017：764-773.
  13. [13] ZHENG Z H，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.AAAI：New York， 2020，34（7）：12993-13000.
  14. [14] ZHANG Y F，REN W Q，ZHANG Z，et al. Focal and Efficient IOU Loss for Accurate Bounding Box Regression [J].Neurocomputing，2022，506：146-157.
  15. [15] GEVORGYAN Z. SIoU Loss: More Powerful Learning for Bounding Box Regression [J/OL].arXiv:2205.12740 [cs.CV].（2022-05-25）.https://arxiv.org/abs/2205.12740.
  16. [16] MA S L，XU Y. MPDIoU: A Loss for Efficient and Accurate Bounding Box Regression [J/OL].arXiv:2307.07662 [cs.CV].（2023-07-14）.https://arxiv.org/abs/2307.07662.
  17. [17] TONG Z J，CHEN Y H，XU Z W，et al. Wise-IoU: Bounding Box Regression Loss with Dynamic Focusing Mechanism [J/OL].arXiv:2301.10051 [cs.CV].（2023-04-08）.https://arxiv.org/abs/2301.10051.

About the authors

Han Weijuan (1986-), female, Han ethnicity, from Yuzhou, Henan Province, is an intermediate engineer with a master's degree. Her research interests include deep learning, image processing, and wireless sensor networks.