Identification of dead trees using artificial intelligence based on Earth remote sensing data
https://doi.org/10.25587/2587-8751-2024-4-138-149
Abstract
On the territory of the Russian Federation, there are massive shrinking of the stand, which are caused by the effects of pathogens, insects, fungal infestations, as well as the spread of bacterial diseases of trees. The development of remote monitoring systems for forest ecosystems is extremely important for the management of forest resources and making informed decisions regarding the conservation and restoration of forests. The purpose of this study was the development of a software module for automating the identification process of dead trees based on Earth remote sensing data. The analysis of various detection and segmentation strategies, including traditional computer vision methods and neural networks based on artificial intelligence, made it possible to choose object detection as the main method due to its effectiveness in labeling and the ability to quantify areas of tree damage. The combination of the object detection method with high-resolution images obtained using UAVs proved to be the most effective for accurate detection of dead trees.
Keywords
artificial intelligence,
climate change,
remote sensing of the Earth,
unmanned aerial vehicles,
satellite imagery,
object detection,
segmentation
Supporting agencies: The research was carried out within the framework of the program of the Ministry of Science and Higher Education of the Russian Federation “Priority 2030” (National project “Science and Universities”).
About the Authors
R. D. Shagaliev
Ufa State Petroleum Technological University
Russian Federation
Russian Federation
Shagaliev Ruslan D. – Candidate of Technical Sciences, head of the Interuniversity Laboratory of Climate and Carbon Footprint Monitoring
Ufa
E. A. Bogdan
Ufa State Petroleum Technological University
Russian Federation
Russian Federation
Bogdan Ekaterina A. – Candidate of Economic Sciences, Leading Researcher
Ufa
A. F. Galyamov
Ufa State Petroleum Technological University
Russian Federation
Russian Federation
Galyamov Airat F. – Candidate of Technical Sciences, Associate Professor
Ufa
L. N. Belan
Ufa State Petroleum Technological University
Russian Federation
Russian Federation
Belan Larisa N. – Doctor of Geological and Mineralogical Sciences, Director of the Center for Decarbonization Technologies
Ufa
O. I. Ishkinina
Ufa State Petroleum Technological University
Russian Federation
Russian Federation
Ishkinina Olesya I. – Candidate of Chemical Sciences, Associate professor
Ufa
G. G. Valiev
Ufa State Petroleum Technological University
Russian Federation
Russian Federation
Valiev Gaziz G. – laboratory engineer
Ufa
References
1. Pyzhev AI, Syrtsova EA, Zander EV. Forest resources of Asian Russia: wealth or scarcity? Journal of Siberian Federal University. Humanities & Social Sciences, 2022;15 (12):1841-1853. In Russia
2. Malakhova EG, Lyamtsev NI. Extent and structure of Moscow region spruce forest dieback in 2010-2012. Proceedings of the St. Petersburg Forestry Engineering Academy, 2014;207:193-201. In Russia
3. Mirtova IA, Ershov DV. The use of satellite data to assess damage to forests by bark beetles on the example of the Moscow region. News of higher educational institutions. Geodesy and aerial photography, 2013;6: 7-82. In Russia
4. Gninenko YuI, Ivanov VA. The possibilities of using satellite images to track the foci of the Allied bark beetle. Actual problems of the forest complex, 2022;62:137-142. In Russia
5. Karkhova SA, Nikitenko EB. Assessment of the forest-pathological condition of cedar forests of Irkutsk region. Bulletin of Baikal state university, 2023;33(2):80-393. In Russia
6. Koltunov EV, Erdakov LN. Spectral analysis of the long-term dynamics of outbreaks of mass reproduction of the unpaired silkworm (Lymantria dispar L.) in the Urals. Modern problems of science and education, 2013;2:399. In Russia
7. Maslov AD. Reproduction of spruce stem pests in foci of root rot. Forest protection from pests and diseases: proceedings of VNIILM. Moscow: Forest Industry, 1973:84–10. In Russia
8. Kolganikhina GB, Sinkevich VV. To the study of the problem of drying of elms in Moscow and the Moscow region. Proceedings of the Saint Petersburg forestry research institute, 2021;3:67-85. In Russia
9. Selikhovkin AV, Glebov RV, Magdeev NG, et al. Assessment of the role of insects and dendropathogenic organisms in the drying of stands of the Leningrad region and the Republic of Tatarstan. Forestry, 2016.;2:83-95. In Russia
10. Singatullin IK. The state of the aspen trees of the Republic of Tatarstan after the drought of 2010. Bulletin of the Kazan State Agrarian University, 2016;3(41):40-45. In Russia
11. Bogdan EA. Kamalova RG, Belan LN, Tuktarova IO. The influence of climatic changes on the spread of bacterial dropsy of birch. Geographical Bulletin, 2024;1(68):151-165. In Russia
12. Bondarenko-Borisova IV. Bacterial dropsy – a dangerous disease of birch in the Donetsk region. Industrial Botany, 2020;20(2):62-65. In Russia
13. Alekseev AS, Chernikhovsky DM. Identification of the early stages of damage to spruce stands by bark beetles based on a combined analysis of Sentinel-2b satellite imagery and ground surveys. Proceedings of the Saint Petersburg Forestry Research Institute, 2023;246:22-43. In Russia
14. Dolgacheva LE, Rotanova IN. Assessment of damage to forest plantations of the Gorno-Kolyvan forestry of the Altai territory by the polygraphus proximus blandford using remote sensing data. Advances in current natural sciences, 2023;7:21-26. In Russia
15. Sannikov IYu, Andreev DN, Buzmakov SA. Identification and analysis of deadwood using an unmanned aerial vehicle. Сosmic Research, 2018;15(3):103-113. In Russia
16. Ivanova NV, Shashkov MP, Shanin VN. Obtaining tree stand attributes from unmanned aerial vehicle (UAV) data: the case of mixed forests. Tomsk State University Journal of Biology, 2021,54:158-175. In Russia
17. Safonova A, Hamad Y, Alekhina A, Kaplun D. Detection of Norway spruce trees (Picea Abies) infested by bark beetle in UAV images using YOLOs architectures. IEEE Access, 2022,10:10384-10392.
18. Safonova A, Tabik S, Alcaraz-Segura D [et al.] Detection of fir trees (Abies sibirica) damaged by the bark beetle in unmanned aerial vehicle images with deep learning. Remote Sensing, 2019,11(6):643.
19. Safonova A., Hamad Y., Dmitriev E. [et al.] Individual tree crown delineation for the species classification and assessment of vital status of forest stands from UAV images. Drones, 2021,5(3):77.
Review
For citations:
Shagaliev R.D., Bogdan E.A., Galyamov A.F., Belan L.N., Ishkinina O.I., Valiev G.G. Identification of dead trees using artificial intelligence based on Earth remote sensing data. Vestnik of North-Eastern Federal University Series "Earth Sciences". 2024;(4):138-149. (In Russ.) https://doi.org/10.25587/2587-8751-2024-4-138-149
Views:
140
Email this article 