Affiliation:
1. Southwest State University
Abstract
The purpose of the research – analysis of the effectiveness of two-stage neural network models for solving the problem of detecting forest fires in images obtained from unmanned aerial vehicles.Methods. А training dataset was synthesized for training neural network models for the purpose of detection and semantic segmentation of forest fires in images. Тwo-stage neural network models (“Faster R-CNN”, “Mask RCNN” and “Retina-Net”) were used for training. Тhe neural network models were trained according to the same parameters set for all models in order to ensure consistency and a common basis for experiments. Optimization of model parameters during the training process was carried out to minimize the classification loss function. Тo synthesize the test sample, we used a video sequence covering the events of forest fires in the /rkutsk region, which was filmed by an unmanned aerial vehicle. Using a specially developed script in the Рython programming language, the process of dividing this video sequence into separate frames was carried out, which were used as a test data set when assessing the quality of classification of trained neural network models.Results. Based on the analysis of the obtained values of the quality criterion, as well as visual analysis on the test data set produced as part of testing neural network models, the effectiveness of the studied models for detecting forest fires in images was assessed. Тo assess the quality of binary classification of neural network models, the quality criterion “Accuracy” (classification accuracy) was used.Conclusion. Еxperimental studies on a test data set showed that the Retina-Net model demonstrates the lowest, but acceptable, performance compared to other studied neural network models. Тhe two-stage neural network models “Faster R-CNN” and “Mask R-CNN” demonstrate similar classification accuracy values (0.9492 and 0.9521, respectively), which allows us to recommend them for use in early detection systems for forest fires.
Publisher
Southwest State University
Reference21 articles.
1. Sibirkina A. R., Lihachev S. F. O esnyh pozharah v esah Che yabins oj ob asti za 2018-2021 gody i ana iz trebovanij vosproizvodstvu esov v esohozyajstvennom reg amente [On forest fires in the forests of the Che yabins region for 2018–2021 and ana ysis of requirements for forest reproduction in forest management regu ations]. Lesnoj vestnik = Forest Bulletin, 2023, vo . 27, no. 5, pp. 60–73. https://doi.org/10.18698/2542-1468-2023-5-60-73
2. Yang L., Yusheng S. Estimates of G oba Forest Fire Carbon Emissions Using FY-3 Active Fires Product. Atmosphere, 2023, vo . 14, p. 1575. https://doi.org/10.3390/atmos14101575
3. Luz V., eds. Eco ogica Impact of Forest Fires and Subsequent Restoration in Chi e. Resources, 2018, vo . 7, no. 2, p. 26. https://doi.org/10.3390/resources7020026
4. Konkova Yu. M. Lesnyye pozhary, a fa tor utraty fun tsiy esny h e osistem [Forest fires as a factor in the oss of functions of forest ecosystems]. Novosti nauki v APК = Science News in the Agro-lndustrial Complex, 2021, vo . 2, pp. 111–113. https://doi.org/10.25930/2218-855x/029.2.2021
5. Kovalev R. N., Ena eeva-Bandura I. M., Ni onchu A. V. Ocen a v iyaniya pozharov na esnye e osistemy C uchetom urovnya razvitiya esotransportnoj seti [Assessment of fire impact on forest ecosystems ta ing into account the eve of forest transportation networ deve opment]. lzvestiуa vuzov. Lesnoj zhurnal = Proceedings rom Universities. Forestrу Journal, 2021, vo . 4, pp. 131–49. https://doi.org/10.37482/0536-1036-2021-4-131-149