Neural Network Methods for Detecting Fires in Forests

Vitaly P. Fralenko; Фраленко Виталий Петрович

doi:10.14357/20718594230107

Neural Network Methods for Detecting Fires in Forests

Autores: Fralenko V.P.¹
Afiliações:
1. A. K. Ailamazyan Program Systems Institute of the Russian Academy of Sciences
Edição: Nº 1 (2023)
Páginas: 67-77
Seção: Machine Learning, Neural Networks
URL: https://bakhtiniada.ru/2071-8594/article/view/269811
DOI: https://doi.org/10.14357/20718594230107
ID: 269811

Citar

Resumo

This work includes an analytical review, investigated, supplemented and tested actual neural network methods, algorithms and approaches for solving the problem of early detection of fires in forests using images and video streams from unmanned aerial vehicles. The proposed scheme for solving the problem is based on feature extraction and the use of machine learning for frame classification, selection of a rectangular region with target fire sources and accurate semantic segmentation of fires using convolutional neural networks. The performed modifications of the architectures of neural networks are described, which made it possible to improve the F1-measures achieved by them by 20%.

Palavras-chave

fire monitoring, unmanned aerial vehicle, neural network, frame classification, localization of target areas of interest, semantic segmentation

Texto integral

Sobre autores

Vitaly Fralenko

A. K. Ailamazyan Program Systems Institute of the Russian Academy of Sciences

Autor responsável pela correspondência
Email: alarmod@pereslavl.ru

Candidate of technical sciences. Leading researcher

Rússia, Veskovo, Yaroslavl region

Bibliografia

Geetha S., Abhishek C.S., Akshayanat C.S. Machine Vision Based Fire Detection Techniques: A Survey. Fire Technology, 2021, Vol.57, pp.591-623.
Ciprián-Sánchez J.F., Ochoa-Ruiz G., Rossi L. and etc. “Assessing the Impact of the Loss Function, Architecture and Image Type for Deep Learning-Based Wildfire Segmentation”. Appl. Sci., 2021, Vol.11(15), 7046.
Cruz H., Gualotuña T., Pinillos M. and etc. “Machine Learning and Color Treatment for the Forest Fire and Smoke Detection Systems and Algorithms, a Recent Literature Review”. In: Botto-Tobar, M., Cruz, H., Díaz Cadena, A. (eds) Artificial Intelligence, Computer and Software Engineering Advances. CIT 2020. Advances in Intelligent Systems and Computing, 2021, Vol.1326. Springer, Cham.
Shubhangi Chaturvedi, Pritee Khanna, Aparajita Ojha “A Survey on Vision-Based Outdoor Smoke Detection Techniques for Environmental Safety”. ISPRS Journal of Photogrammetry and Remote Sensing, 2022, Vol.185, pp.158-187.
Favorskaya M.N. Early Smoke Detection in Outdoor Space: State-of-the-Art, Challenges and Methods. In: Virvou, M., Tsihrintzis, G.A., Jain, L.C. (eds). Advances in Selected Artificial Intelligence Areas. Learning and Analytics in Intelligent Systems, 2022, Vol.24. Springer, Cham.
Cheng Y., Chen K., Bai H. and etc. “An Efficient Fire Detection Algorithm Based on Multi-Scale Convolutional Neural Network”. Fire and Materials, 2021, 12 p.
Huo Y., Zhang Q., Jia Y. and etc. A Deep Separable Convolutional Neural Network for Multiscale Image-Based Smoke Detection // Fire Technology, 2022.
Miao J., Zhao G., Gao Y. and etc. Fire Detection Algorithm Based on Improved YOLOv5 // International Conference on Control, Automation and Information Sciences (ICCAIS). 2021. Р. 776-781.
Li Y., Zhang W., Liu Y. and etc. “A Visualized Fire Detection Method Based on Convolutional Neural Network Beyond Anchor”. Applied Intelligence, 2022.
Wang S., Zhao J., Ta N. and etc. “A Real-Time Deep Learning Forest Fire Monitoring Algorithm Based on an Improved Pruned + Kd Model”. J Real-TimeImage Proc 18, 2021, pp.2319-2329.
Wang S. and etc. “Forest Fire Detection Based on Light-weight Yolo” 2021 33rd Chinese Control and Decision Conference (CCDC), 2021, pp.1560-1565.
Yiheng Cai, Yajun Guo, Yuanyuan Li and etc. “Fire Detection Method Based on Improved Deep Convolution Neural Network”. In Proceedings of the 2019 8th International Conference on Computing and Pattern Recognition (ICCPR '19), 2019. Association for Computing Machinery, New York, NY, USA, pp.466-470.
Athanasis N., Themistocleous M., Kalabokidis K. and etc. “Big Data Analysis in UAV Surveillance for Wildfire Prevention and Management”. Themistocleous, M., Rupinoda Cunha, P. (eds) Information Systems. EMCIS 2018. Lecture Notes in Business Information Processing, 2019, Vol.341. Springer, Cham.
Ghali R., Akhloufi M.A., Mseddi W.S. Deep Learning and Transformer Approaches for UAV-Based Wildfire Detection and Segmentation. Sensors, 2022, Vol.22(5), 1977.
Wu B., Xu C., Dai X. and etc. Visual Transformers: Token-based Image Representation and Processing for Computer Vision, 2020. URL: https://arxiv.org/abs/2006.03677 (last visited: 27.10.2022).
Shamsoshoara A., Afghah F., Razi A. and etc. “The Flame Dataset: Aerial Imagery Pile Burn Detection Using Drones (UAVS)”, 2021. URL: https://ieee-dataport.org/open-access/flame-dataset-aerial-imagery-pile-burn-detection-using-drones-uavs (last visited: 27.10.2022).
GitHub – WongKinYiu/PyTorch_YOLOv4: PyTorch implementation of YOLOv4, 2021. URL: https://github.com/WongKinYiu/PyTorch_YOLOv4 (last visited: 27.10.2022).
GitHub – ultralytics/yolov5: YOLOv5 in PyTorch > ONNX CoreML > TFLite, 2022. URL:https://github.com/ultralytics/yolov5/ (last visited: 27.10.2022).
GitHub – meituan/YOLOv6: YOLOv6: a single-stage object detection framework dedicated to industrial applications, 2022. URL: https://github.com/meituan/YOLOv6 (last visited: 27.10.2022).
GitHub – WongKinYiu/yolov7: Implementation of paper – YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors, 2022. URL: https://github.com/WongKinYiu/yolov7 (last visited: 27.10.2022).
Image classification from scratch, 2020. URL: https://keras.io/examples/vision/image_classification_fro m_scratch/ (last visited: 27.10.2022).
Thomson W., Bhowmik N., Breckon T.P. Efficient and Compact Convolutional Neural Network Architectures for Non-temporal Real-time Fire Detection, 2020. URL: https://arxiv.org/pdf/2010.08833.pdf (last visited: 27.10.2022).
Akagic A., Buza E. LW-FIRE: A Lightweight Wildfire Image Classification with a Deep Convolutional Neural Network. Appl. Sci., 2022, Vol.12(5), 2646.
Public API for tf.keras.applications namespace, 2022. URL: https://www.tensorflow.org/api_docs/python/tf/keras/appli cations (last visited: 27.10.2022).
Fralenko V.P. Experimental Investigation with «Darknet» Neural Network Capabilities on the Task of Remote Sensing Images Processing // Aerospace Instrument-Making. 2017. No.6. pp.44-52.
Yaowen Hu, Jialei Zhan, Guoxiong Zhou and etc. “Fast Forest Fire Smoke Detection Using MVMNet”. Knowledge-Based Systems, 2022, Vol.241, 20 p. doi: 10.1016/j.knosys.2022.108219.
Ronneberger O., Fischer P., Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation, 2015.
Vahit Bugra Yesilkaynak, Yusuf H. Sahin, Gozde Unal “EfficientSeg: An Efficient Semantic Segmentation Network”, 2020.
Hu Cao, Yueyue Wang, Joy Chen and etc. “Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation”, 2021.
Bhakti Baheti, Shubham Innani, Suhas Gajre and etc. “Eff-UNet: A Novel Architecture for Semantic Segmentation in Unstructured Environment”, 2020. URL: https://openaccess.thecvf.com/content_CVPRW_2020/papers/w22/Baheti_Eff-UNet_A_Novel_Architecture_for_Semantic_Segmentation_in_Unstructured_Environment_CVPRW_2020_paper.pdf (last visited: 27.10.2022).
Abhishek Chaurasia, Eugenio Culurciello “LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation”, 2017.