ABSTRACT
Deep learning models, such as YOLOv5, well-known for object detection, and U-Net, used for segmentation, are known for their respective capabilities within computer vision tasks. In this study, the researchers introduced a novel framework that uses YOLOv5 and U-Net models, in combination with frame differencing techniques, to achieve early fire detection in an indoor setting. YOLOv5 was trained on a diverse dataset consisting of fire, smoke, and non-fire scenarios, while U-Net was exclusively trained on fire data. Motion detection was then implemented using frame differencing that allowed to identify fire movements effectively. The developed framework achieved an overall accuracy of 88%, outperforming the standalone YOLOv5 model with its 81% accuracy. This improvement of 7% in detection performance was influenced by the incorporation of fire motion analysis which effectively reduced false positive results. In summary, the study presents a robust framework that significantly improves fire detection in indoor environments with the help of motion analysis alongside the used deep learning models.
- Mseddi, W.S., et. al. 2021. Fire Detection and Segmentation using YOLOv5 and U-NET. (2021). DOI: http://dx.doi.org/10.23919/EUSIPCO54536.2021.9616026Google ScholarCross Ref
- Shiwei Wang, Feng Yu, Changlong Zhou, and Minghua Jiang. 2020. Straw burning detection method based on improved frame difference method and deep learning. 2020 IEEE 5th International Conference on Image, Vision and Computing (ICIVC) (2020). DOI: http://dx.doi.org/10.1109/icivc50857.2020.9177456Google ScholarCross Ref
- Kumar, K., Varma, K., Sujihelen, L., Jancy, S., Aishwarya, R. and Yogitha, R., 2022. Computer Vision-Based Early Fire Detection Using Machine Learning. 2022 International Conference on Communication, Computing and Internet of Things (IC3IoT). DOI: http://dx.doi.org/10.1109/IC3IOT53935.2022.9767886Google ScholarCross Ref
- Sthevanie, F., Kosala, G., Ramadhani, K.N., and Putra, K.S. (2021) Fire Detection on Video Using ViBe Algorithm and LBP-TOP. Jurnal RESTI: Rekayasa Sistem dan Teknologi Informasi. DOI: https://doi.org/10.29207/resti.v6i6.4164Google ScholarCross Ref
- Ryu, J. and Kwak, D., 2021. Flame Detection Using Appearance-Based Pre-Processing and Convolutional Neural Network. Applied Sciences, [online] 11(11), p.5138. DOI: http://dx.doi.org/10.3390/app11115138Google ScholarCross Ref
- Zhang, B., Sun, L., Song, Y., Shao, W., Guo, Y. and Yuan F. (2020) DeepFireNet: A real-time video fire detection method based on multi-feature fusion. Aimpress. Mathematical Biosciences and Engineering. DOI: 10.3934/mbe.2020397Google ScholarCross Ref
- Hongyu, H., Ping, K., Fan, L., Huaxin, S. 2020. An Improved Multi-Scale Fire Detection Method based on Convolutional Neural Network. (2020). DOI: http://dx.doi.org/10.1109/ICCWAMTIP51612.2020.9317360Google ScholarCross Ref
- An, Q., Chen, X., Zhang, J., Shi, R., Yang, Y. and Huang, W., 2022. A Robust Fire Detection Model via Convolution Neural Networks for Intelligent Robot Vision Sensing. Sensors, [online] 22(8), p.2929. DOI: http://dx.doi.org/10.3390/s22082929Google ScholarCross Ref
- Mwedzi N.A., Nwulu N.I., and Gbadamosi S.L. 2019. Machine Learning Applications for Fire Detection in a Residential Building. IEEE. [Online]. DOI: http://dx.doi.org/10.1109/ICETAS48360.2019.9117318Google ScholarCross Ref
- Chen, K., et. al. 2019. Research on Image Fire Detection Based on Support Vector Machine. (2019). DOI: http://dx.doi.org/10.1109/ICFSFPE48751.2019.9055795Google ScholarCross Ref
- K. Chen, Y. Cheng, H. Bai, C. Mou and Y. Zhang, "Research on Image Fire Detection Based on Support Vector Machine," 2019 9th International Conference on Fire Science and Fire Protection Engineering (ICFSFPE), Chengdu, China, 2019, pp. 1-7, doi: 10.1109/ICFSFPE48751.2019.9055795Google ScholarCross Ref
- Guohua Wang, Juncong Li, Yongsen Zheng, Qi Long, and Wanrong Gu. 2020. Forest smoke detection based on Deep Learning and background modeling. 2020 IEEE International Conference on Power, Intelligent Computing and Systems (ICPICS) (2020). DOI: http://dx.doi.org/10.1109/icpics50287.2020.9202287Google ScholarCross Ref
- Gonçalo Perrolas, Milad Niknejad, Ricardo Ribeiro, and Alexandre Bernardino. 2022. Scalable fire and smoke segmentation from aerial images using convolutional neural networks and quad-Tree Search. Sensors 22, 5 (2022), 1701. DOI:http://dx.doi.org/10.3390/s22051701Google ScholarCross Ref
- Xue, Z., Lin, H. and Wang, F., 2022. A Small Target Forest Fire Detection Model Based on YOLOv5 Improvement. Forests, [online] 13(8), p.1332. DOI: http://dx.doi.org/10.3390/f13081332Google ScholarCross Ref
- Hurley, Morgan J. and Richard W. Bukowski. “Fire Hazard Analysis Techniques.” (2019).Google Scholar
- Sida Dai, Tao Han, Mingchuan Yang, Yuan Zhang, and Lichuan Wang. 2022. Towards Small Scale Urban Fire Detection Dataset. 2022 IEEE 9th International Conference on Cyber Security and Cloud Computing (CSCloud)/2022 IEEE 8th International Conference on Edge Computing and Scalable Cloud (EdgeCom) (2022). DOI:http://dx.doi.org/10.1109/cscloud-edgecom54986.2022.00026Google ScholarCross Ref
- Yuanpan Zheng, Zhenyu Wang, Boyang Xu, and Yiqing Niu. 2022. Multi-scale semantic segmentation for fire smoke image based on global information and U-Net. Electronics 11, 17 (2022), 2718. DOI: http://dx.doi.org/10.3390/electronics11172718Google ScholarCross Ref
- Yafei Gong, Xuanchao Ma, Hongyan Liu, and Ke Gu. 2022. Multiple categories of visual smoke detection database. 2022 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS) (2022). DOI:http://dx.doi.org/10.1109/ispacs57703.2022.10082827Google ScholarCross Ref
- Yessi Jusman, Lentera Hinggis, Rama Okta Wiyagi, Nor Ashidi Mat Isa, and Faaris Mujaahid. 2020. Comparison of Background Subtraction and Frame Differencing Methods for Indoor Moving Object Detection. 2020 1st International Conference on Information Technology, Advanced Mechanical and Electrical Engineering (ICITAMEE) (2020). DOI:https://doi.org/10.1109/icitamee50454.2020.9398484Google ScholarCross Ref
- Faming Gong 2019. A real-time fire detection method from video with Multifeature Fusion. Computational Intelligence and Neuroscience 2019 (2019), 1–17. DOI:http://dx.doi.org/10.1155/2019/1939171Google ScholarDigital Library
Index Terms
- Early Fire Detection and Segmentation Using Frame Differencing and Deep Learning Algorithms with an Indoor Dataset
Recommendations
Early Fire Detection and Alert System using Modified Inception-v3 under Deep Learning Framework
AbstractADSI-2019 report states that there were approximately eleven thousand fire accidents reported across India in 2019. In this article, a deep learning model has been developed for fire and smoke detection from images. In this context, we have ...
Semi-supervised Deep Learning for Network Anomaly Detection
Algorithms and Architectures for Parallel ProcessingAbstractDeep learning promotes the fields of image processing, machine translation and natural language processing etc. It also can be used in network anomaly detection. In practice, it is not hard to obtain normal instances. However, it is always ...
Comments