Fire is a common but serious disaster, which poses a great threat to human life and property. Therefore, fire-smoke detection technology is of great significance in various fields. In order to improve the detection ability of tiny-fire, so as to realize the prediction and suppression of fire as soon as possible, we proposed an efficient and accurate tiny-fire detection method based on the optimized YOLOv7, and we named the improved model YOLOv7-FIRE. First, we introduced the BiFormer into YOLOv7 to make the network pay more attention to the fire-smoke area. Second, we introduced the NWD technique to enhance the perception of the algorithm for small targets, and provided richer semantic information by modeling the context information around the target. Finally, CARAFE was applied for content-aware feature reorganization, which preserved the details and texture information in the image and improved the quality of fire-smoke detection. Furthermore, in order to improve the robustness of the improved algorithm, we expanded the fire-smoke dataset. The experimental results showed that YOLOv7-FIRE as significantly better than the previous algorithm in detection accuracy and recall rate, the Precision increased from 75.83% to 82.31%, and the Recall increased from 66.43% to 74.02%.
Citation: Baoshan Sun, Kaiyu Bi, Qiuyan Wang. YOLOv7-FIRE: A tiny-fire identification and detection method applied on UAV[J]. AIMS Mathematics, 2024, 9(5): 10775-10801. doi: 10.3934/math.2024526
Fire is a common but serious disaster, which poses a great threat to human life and property. Therefore, fire-smoke detection technology is of great significance in various fields. In order to improve the detection ability of tiny-fire, so as to realize the prediction and suppression of fire as soon as possible, we proposed an efficient and accurate tiny-fire detection method based on the optimized YOLOv7, and we named the improved model YOLOv7-FIRE. First, we introduced the BiFormer into YOLOv7 to make the network pay more attention to the fire-smoke area. Second, we introduced the NWD technique to enhance the perception of the algorithm for small targets, and provided richer semantic information by modeling the context information around the target. Finally, CARAFE was applied for content-aware feature reorganization, which preserved the details and texture information in the image and improved the quality of fire-smoke detection. Furthermore, in order to improve the robustness of the improved algorithm, we expanded the fire-smoke dataset. The experimental results showed that YOLOv7-FIRE as significantly better than the previous algorithm in detection accuracy and recall rate, the Precision increased from 75.83% to 82.31%, and the Recall increased from 66.43% to 74.02%.
| [1] |
F. Q. Zhang, P. C. Zhao, S. W. Xu, Y. Wu, X. B. Yang, Y. Zhang, Integrating multiple factors to optimize watchtower deployment for wildfire detection, Sci. Total Environ., 737 (2020), 139561. https://doi.org/10.1016/j.scitotenv.2020.139561 doi: 10.1016/j.scitotenv.2020.139561
|
| [2] | M. Karthi, R. Priscilla, S. G, N. Infantia C, A. G. R, V. J, Forest fire detection: A comparative analysis of deep learning algorithms, 2023 International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent Engineering (ICECONF), Chennai, India, 2023, 1–6. https://doi.org/10.1109/ICECONF57129.2023.10084329 |
| [3] | J. Redmon, S. Divvala, R. Girshick, A. Farhadi, You only look once: Unified, real-time object detection, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016,779–788. https://doi.org/10.1109/CVPR.2016.91 |
| [4] | J. Redmon, A. Farhadi, Yolo9000: better, faster, stronger, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017, 6517–6525. https://doi.org/10.1109/CVPR.2017.690 |
| [5] |
Z. Liu, X. Y. Gu, H. L. Yang, L. T. Wang, Y. H. Chen, D. Y. Wang, Novel YOLOv3 model with structure and hyperparameter optimization for detection of pavement concealed cracks in GPR images, IEEE T. Intell. Transp., 23 (2022), 22258–22268. https://doi.org/10.1109/TITS.2022.3174626 doi: 10.1109/TITS.2022.3174626
|
| [6] | A. Bochkovskiy, C. Y. Wang, H. Y. M. Liao, YOLOv4: Optimal speed and accuracy of object detection, 2020, arXiv: 2004.10934. https://doi.org/10.48550/arXiv.2004.10934 |
| [7] |
Q. Q. Ding, P. Li, X. F. Yan, D. Shi, L. M. Liang, W. M. Wang, et al., CF-YOLO: Cross fusion YOLO for object detection in adverse weather with a high-quality real snow dataset, IEEE T. Intell. Transp., 24 (2023), 10749–10759. https://doi.org/10.1109/TITS.2023.3285035 doi: 10.1109/TITS.2023.3285035
|
| [8] |
K. C. Song, X. K. Sun, S. Ma, Y. H. Yan, Surface defect detection of aeroengine blades based on cross-layer semantic guidance, IEEE T. Instrum. Meas., 72 (2023), 2514411. https://doi.org/10.1109/TIM.2023.3276026 doi: 10.1109/TIM.2023.3276026
|
| [9] | C. Y. Wang, A. Bochkovskiy, H. Y. M. Liao, YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors, 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada, 2023, 7464–7475. https://doi.org/10.1109/CVPR52729.2023.00721 |
| [10] | L. Zhu, X. J. Wang, Z. H. Ke, W. Zhang, R. Lau, BiFormer: Vision transformer with Bi-level routing attention, 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada, 2023, 10323–10333. https://doi.org/10.1109/CVPR52729.2023.00995 |
| [11] | J. W. Wang, C. Xu, W. Yang, L. Yu, A normalized Gaussian Wasserstein distance for tiny object detection, 2021, arXiv: 2110.13389. https://doi.org/10.48550/arXiv.2110.13389 |
| [12] | J. Q. Wang, K. Chen, R. Xu, Z. W. Liu, C. C. Loy, D. H. Lin, CARAFE: Content-aware ReAssembly of Features, 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea (South), 2019, 3007–3016. https://doi.org/10.1109/ICCV.2019.00310 |
| [13] | A. Kirillov, R. Girshick, K. He, P. Dollár, Panoptic Feature Pyramid Networks, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, 2019, 6392–6401. https://doi.org/10.1109/CVPR.2019.00656 |
| [14] | M. Hu, Y. L. Li, L. Fang, S. J. Wang, A2-FPN: Attention aggregation based feature pyramid network for instance segmentation, 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 2021, 15338–15347. https://doi.org/10.1109/CVPR46437.2021.01509 |
| [15] |
Z. Yi, J. Ma, X. H. Li, J. Zhang, Saliency detection and deep learning-based wildfire identification in UAV imagery, Sensors, 18 (2018), 712. https://doi.org/10.3390/s18030712 doi: 10.3390/s18030712
|
| [16] |
X. Q. Li, Z. X. Chen, Q. M. J. Wu, C. Y. Liu, 3D parallel fully convolutional networks for real-time video wildfire smoke detection, IEEE T. Circ. Syst. Vid., 30 (2020), 89–103. https://doi.org/10.1109/TCSVT.2018.2889193 doi: 10.1109/TCSVT.2018.2889193
|
| [17] |
S. Q. Ren, K. M. He, R. Girshick, J. Sun, Faster R-CNN: Towards real-time object detection with region proposal networks, IEEE T. Pattern Anal., 39 (2017), 1137–1149. https://doi.org/10.1109/TPAMI.2016.2577031 doi: 10.1109/TPAMI.2016.2577031
|
| [18] | J. L. Zhang, S. X. Chen, Y. W. Hou, Accurate object detection with relation module on improved R-FCN, 2020 Chinese Automation Congress (CAC), Shanghai, China, 2020, 7131–7135. https://doi.org/10.1109/CAC51589.2020.9326543 |
| [19] |
J. J. Ni, K. Shen, Y. Chen, S. X. Yang, An improved SSD-like deep network-based object detection method for indoor scenes, IEEE T. Instrum. Meas., 72 (2023), 5006915. https://doi.org/10.1109/TIM.2023.3244819 doi: 10.1109/TIM.2023.3244819
|
| [20] |
R. J. Xu, H. F. Lin, K. J. Lu, L. Cao, Y. F. Liu, A forest fire detection system based on ensemble learning, Forests, 12 (2021), 217. https://doi.org/10.3390/f12020217 doi: 10.3390/f12020217
|
| [21] | M. X. Tan, R. M. Pang, Q. V. Le, EfficientDet: Scalable and efficient object detection, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2020, 10778–10787. https://doi.org/10.1109/CVPR42600.2020.01079 |
| [22] |
K. Ramamurthy, A. R. Varikuti, B. Gupta, N. Aswani, A deep learning network for Gleason grading of prostate biopsies using EfficientNet, Biomed. Tech., 68 (2023), 187–198. https://doi.org/10.1515/bmt-2022-0201 doi: 10.1515/bmt-2022-0201
|
| [23] | S. C. Ren, D. Q. Zhou, S. F. He, J. S. Feng, X. C. Wang, Shunted Self-Attention via Multi-Scale Token Aggregation, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA, 2022, 10843–10852. https://doi.org/10.1109/CVPR52688.2022.01058 |
| [24] | Z. J. Tong, Y. H. Chen, Z. Xu, R. Yu, Wise-IoU: Bounding box regression loss with dynamic focusing mechanism, 2023, arXiv: 2301.10051. https://doi.org/10.48550/arXiv.2301.10051 |
| [25] |
H. Y. Peng, S. Q. Yu, A systematic IoU-Related method: Beyond simplified regression for better localization, IEEE T. Image Process., 30 (2021), 5032–5044. https://doi.org/10.1109/TIP.2021.3077144 doi: 10.1109/TIP.2021.3077144
|
| [26] | Z. Gevorgyan, SIoU loss: More powerful learning for bounding box regression, 2022, arXiv: 2205.12740. https://doi.org/10.48550/arXiv.2205.12740 |
| [27] |
Z. H. Zheng, P. Wang, D. W. Ren, W. Liu, R. G. Ye, Q. H. Hu, et al., Enhancing geometric factors in model learning and inference for object detection and instance segmentation, IEEE T. Cybernetics, 52 (2022), 8574–8586. https://doi.org/10.1109/TCYB.2021.3095305 doi: 10.1109/TCYB.2021.3095305
|
Figures(17) / Tables(2)
Baoshan Sun, Kaiyu Bi, Qiuyan Wang. YOLOv7-FIRE: A tiny-fire identification and detection method applied on UAV[J]. AIMS Mathematics, 2024, 9(5): 10775-10801. doi: 10.3934/math.2024526
DownLoad: