Abstract
Fire is one of the major disasters in the world, which seriously endangers the safety of life and property. Effective flame and smoke detection can provide timely warning information for firefighters. Existing flame and smoke detection algorithms are limited by processor performance and cannot operate large deep networks. This paper proposes a lightweight detector called Light-YOLOv4, which considers the balance between performance and efficiency. First, the backbone network CSPDarknet53 is replaced by MobileNetv3. Second, the path aggregation network is changed into bidirectional feature pyramid network (BiFPN) based on the idea of bidirectional cross-scale connections. Third, an efficient feature extraction module named depthwise separable attention module based on depthwise separable convolution and coordinate attention network is construed to replace the 3\(\times \)3 standard convolution of spatial pyramid pooling, BiFPN and YOLO head network. In comparison with YOLOv4, Light-YOLOv4 has only 19.1% of its trainable parameters while almost keeping the same accuracy. By combining multiple tricks, Light-YOLOv4 can achieve a better balance between performance (85.64% mAP) and efficiency (71 FPS), which meets flame and smoke detection tasks’ requirements on the accuracy and real time. Experiments on Nvidia Jetson TX2 further demonstrate that Light-YOLOv4 has good detection performance and speed on embedded scenarios.
Similar content being viewed by others
References
Muhammad, K., Ahmad, J., Baik, S.W.: Early fire detection using convolutional neural networks during surveillance for effective disaster management. Neurocomputing 288, 30–42 (2018). https://doi.org/10.1016/j.neucom.2017.04.083
Uǧur Töreyin, B., Dedeoǧlu, Y., Uǧur, G.A., Çetin, E.: Computer vision based method for real-time fire and flame detection. Signal Pattern Recognit. Lett. 27, 49–58 (2006). https://doi.org/10.1016/j.patrec.2005.06.015
Gunay, O., Toreyin, B.U., Kivanc, K.A., Cetin, E.: Entropy-functional-based online adaptive decision fusion framework with application to wildfire detection in video. IEEE Trans. Image Process. 21(5), 2853–2865 (2012). https://doi.org/10.1109/TIP.2012.2183141
Kuo, J.-Y., Lai, T.-Y., Fanjiang, Y.-Y., Huang, F.-C., Liao, Y.-H.: A behavior-based flame detection method for a real-time video surveillance system. J. Chin. Inst. Eng. 38, 947–958 (2015). https://doi.org/10.1080/02533839.2015.1047797
Shamsoshoara, A., Afghaha, F., Razi, A., Zheng, L., Fulé, P.Z., Blasch, E.: Aerial imagery pile burn detection using deep learning: the FLAME dataset. Comput. Netw. (2021). https://doi.org/10.1016/j.comnet.2021.108001
Wang, Y.E., Wei, G.-Y., Brooks D.: Benchmarking TPU, GPU, and CPU Platforms for Deep Learning. (2019) arXiv:1907.10701
Bochkovskiy, A., Wang, C.-Y., Liao, H.-Y.M.: YOLOv4: Optimal Speed and Accuracy of Object Detection. (2020). arXiv:2004.10934
Tan, M., Pang, R., Le, Q.V.: EfficientDet: scalable and efficient object detection. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 10778–10787 (2020). https://doi.org/10.1109/CVPR.2018.00913
Hongyu, H., Ping, K., Li, F., Huaxin, S.: An improved multi-scale fire detection method based on convolutional neural network. In: 2020 17th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP), pp. 109–112 (2020). https://doi.org/10.1109/CVPR.2018.00913
Chino, D.Y.T., Avalhais, L.P.S., Rodrigues Jr, J.F., Train, A.J.M.: BoWFire: detection of fire in still images by integrating pixel color and texture analysis. In: 2015 28th SIBGRAPI Conference on Graphics, Patterns and Images, 2015 28th SIBGRAPI Conference on Graphics, pp. 95–102 (2015). https://doi.org/10.1109/CVPR.2018.00913
Wang, X., Li, Y., Li, Z.: Research on flame detection algorithm based on multi-feature fusion. In: 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), pp. 184–189 (2020). https://doi.org/10.1109/CVPR.2018.00913
Günay, O., Enis Ç.A.: Real-time dynamic texture recognition using random sampling and dimension reduction. In: 2015 IEEE International Conference on Image Processing (ICIP), pp. 3087–3091 (2015).https://doi.org/10.1109/ICIP.2015.7351371
Yu, N., Chen, Y.: Video flame detection method based on TwoStream convolutional neural network. In: 2019 IEEE 8th Joint International Information Technology and Artificial Intelligence Conference (ITAIC), pp. 482–486 (2019). https://doi.org/10.1109/CVPR.2018.00913
Chaoxia, C., Shang, W., Zhang, F.: Information-guided flame detection based on faster R-CNN. IEEE Access 8, 58923–58932 (2020). https://doi.org/10.1109/CVPR.2018.00913
Pan, H., Badawi, D., Cetin, A.E.: Computationally efficient wildfire detection method using a deep convolutional network pruned via Fourier analysis. Sensors 20(10), 2891 (2020). https://doi.org/10.3390/s20102891
Pan, H., Badawi, D., Zhang, X., Cetin, A.E.: Additive neural network for forest fire detection. Signal Image Video Process. SIViP 14, 675–682 (2020). https://doi.org/10.1007/s11760-019-01600-7
He, K., Zhang, X., Ren, S., Sun, J.: Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Trans. Pattern Anal. Mach. Intell. 37(9), 1904–1916 (2015). https://doi.org/10.1109/TPAMI.2015.2389824
Lin, T.-Y., Dollár, P., Girshick, R., He, K., Hariharan, B., Belongie, S.: Feature pyramid networks for object detection. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 936–944 (2017). https://doi.org/10.1109/CVPR.2017.106
Howard, A.G., Zhu, M., Chen, B., et al.: MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications (2017). arXiv:1704.04861
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., Chen, L.-C.: MobileNetV2: inverted residuals and linear bottlenecks. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4510–4520 (2018). https://doi.org/10.1109/CVPR.2018.00474
Howard, A., Sandler, M., Chu, G., et al.: Searching for MobileNetV3. In: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), pp. 1314–1324 (2019). https://doi.org/10.1109/CVPR.2018.00913
Jie, H., Shen, L., Albanie, S., Sun, G., Enhua, W.: Squeeze-and-excitation networks. IEEE Trans. Pattern Anal. Mach. Intell. 42(8), 2011–2023 (2020). https://doi.org/10.1109/TPAMI.2019.2913372
Hou, Q., Zhou, D., Feng, J.: Coordinate Attention for Efficient Mobile Network Design (2021). arXiv:2103.02907
Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Trans. Anal. Mach. Intell. 39(6), 1137–1149 (2017). https://doi.org/10.1109/CVPR.2018.00913
Liu, W., Anguelov, D., Erhan, D., et al.: SSD: single shot multibox detector. In: European Conference on Computer Vision. Lecture Notes in Computer Science, vol. 9905 (2016)
Redmon, J., Redmon J.: YOLOv3: An Incremental Improvement (2018). arXiv:1804.02767
Muhammad, K., Ahmad, J., Mehmood, I., Rho, S., Sung, W.B.: Convolutional neural networks based fire detection in surveillance videos. IEEE Access 6, 18174–18183 (2018). https://doi.org/10.1109/ACCESS.2018.2812835
Muhammad, K., Ahmad, J., Lv, Z., Bellavista, P., Yang, P., Baik, S.W.: Efficient deep CNN-based fire detection and localization in video surveillance applications. IEEE Trans. Syst. Man Cybern. Syst. 49(7), 1419–1434 (2019). https://doi.org/10.1109/TSMC.2018.2830099
Acknowledgements
This work was supported by the National Key R & D Program of China (2019YFB1312104) and Key R & D Program of Hebei Province (20311803D).
Author information
Authors and Affiliations
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, Y., Hua, C., Ding, W. et al. Real-time detection of flame and smoke using an improved YOLOv4 network. SIViP 16, 1109–1116 (2022). https://doi.org/10.1007/s11760-021-02060-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-021-02060-8