Abstract
Identifying a fire in its early stages is essential for minimizing fire incidents by solving it. To prevent from the extent of the fire area, we need effective technology to detect the fire. In this paper, the network structure is slightly improved and compared with the original yolov4 algorithm to explore the features of fire in images to accurately detect fire and realize the fire detection in different scenes. The paper first prepares a variety of fire datasets in complex scenes and divide them into two categories, regular fire and irregular fire, and then label them with labeling software; The labeled datasets are divided into training set and testing set according to 8:2; And then, k-means clustering algorithm is used to initialize the region proposal of the datasets; After that, the origin YOLOv4 and improved model are used to train the training set images respectively under Win10 system; Finally, the trained model is used for the evaluation of testing set and tested on fires in real environment. The experimental results show that the model is effective in detecting fires for various scenes.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Chen, T.H., Wu, P.H., Chiou, Y.C.: An early fire detection method based on i-mage processing. In: Proceedings of the International Conference on Image Processing, Singapore, pp. 1707–1710 (2004)
Celik, T., Demirel, H., Ozkaramanli, H.: Fire detection using statistical color model in video sequences. J. Vis. Commun. Image Represent. 18(2), 176–185 (2007)
Chunyu, Y., Jun, F., Jinjun, W., et al.: Video fire smoke detection using motion and color features. Fire Technol. 46, 651–663 (2010). https://doi.org/10.1007/s10694-009-0110-z
Zaidi, N., Lokman, N., Daud, M.: Fire recognition using RGB and YCbCr color space. ARPN J. Eng. Appl. Sci. 10, 9786–9790 (2015)
Zhaochun, L., Kai, Z.: Research on the identification method for the forest fire based on deep learning. Optik, 223, 165491 (2020)
Wu, S., Zhang, L.: Using popular object detection methods for real time forest fire detection. ISCID 280–284 (2018)
Qixing, Z., Gaohua, L.: Wildland forest fire smoke detection based on faster R-CNN using synthetic smoke images. Proc. Eng. 441–446 (2018)
Redmon, J., Divvala, S., Girshick, R.: You only look once: unified, real time object detection. IEEE CVPR (2016)
Redmon, J., Farhadi, A.:YOLO9000: better, faster, stronger. IEEE CVPR Honolulu 6517–6525 (2017)
Redmon, J., Farhadi, A.: YOLOv3: an incremental improvement. Comput. Vis. Pattern Recognit. 276(7), 126–134 (2018)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Wang, W., Wang, Q., Chen, Y. (2022). Fire Detection Based on YOLOv4 Baseline. In: Zhao, P., Ye, Z., Xu, M., Yang, L., Zhang, L., Yan, S. (eds) Interdisciplinary Research for Printing and Packaging. Lecture Notes in Electrical Engineering, vol 896. Springer, Singapore. https://doi.org/10.1007/978-981-19-1673-1_23
Download citation
DOI: https://doi.org/10.1007/978-981-19-1673-1_23
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-1672-4
Online ISBN: 978-981-19-1673-1
eBook Packages: EngineeringEngineering (R0)