Felling outturn assessment using Earth remote sensing data

A.V. Starovoytov ^a*, A.V. Fattakhov ^a**, E.A. Yachmeneva ^a***, M.M. Khamiev ^a****, D.A. Kisler ^b*****, V.E. Kosarev ^a******, D.K. Nurgaliev ^a*******

^aKazan Federal University, Kazan, 420008 Russia

^b“TNG-Group” Ltd., Bugulma, 423236 Russia

E-mail: ^*aldanstar@gmail.com, ^**avfattahov@kpfu.ru, ^***EAYachmenjova@gmail.com, ^****khamiev@inbox.ru, ^*****denis-kisler@tng.ru, ^******Victor.Kosarev@kpfu.ru, ^*******Danis.Nourgaliev@kpfu.ru

Received October 7, 2021

ORIGINAL ARTICLE

Full text PDF

DOI: 10.26907/2542-064X.2021.4.591-602

For citation: Starovoytov A.V., Fattakhov A.V., Yachmeneva E.A., Khamiev M.M., Kisler D.A., Kosarev V.E., Nurgaliev D.K. Felling outturn assessment using Earth remote sensing data. Uchenye Zapiski Kazanskogo Universiteta. Seriya Estestvennye Nauki, 2021, vol. 163, no. 4, pp. 591–602. doi: 10.26907/2542-064X.2021.4.591-602. (In Russian)

Abstract

Seismic exploration often demands forest clearing, thus making it important to assess the number of trees that must be cut down as the fieldwork proceeds.

We suggest that remote sensing of the Earth’s surface with unmanned aircraft vehicles can be considered as a new approach to solving this problem. To test its validity and potential utility, we installed a laser scanning system and a high-resolution camera on the unmanned aircraft vehicle. The data obtained were used to derive the digital terrain and elevation models of the area under study.

The resulting models were processed with the help of a neural network developed as part of this work. They proved to be useful in identifying trees and their classes within the forest sites subjected to clearing. Additionally, a special algorithm was proposed and applied to assess the felling outturn for each tree class taken separately.

Keywords: unmanned aircraft vehicle, seismic exploration, remote sensing of the Earth, forest clearance

Acknowledgements. This study was funded by the Ministry of Science and Higher Education of the Russian Federation (agreement no. 075-11-2019-038 of November 26, 2019 “Development of a multifunctional hardware and software complex based on unmanned aerial vehicles for planning and support of seismic exploration”).

Figure Captions

Fig. 1. Example of a digitized forest stock table.

Fig. 2. Basic block diagram of processing.

Fig. 3. Structure of the used neural network [10].

Fig. 4. Predicted clearings contoured on the orthophotomap.

Fig. 5. Tree classes shown on the orthophotomap.

References

1. Groshev B.I., Sinitsyn S.G., Moroz P.I., Seperovich I.P. Lesotaksatsionnyi spravochnik [Handbook on Forest Inventory]. Moscow, Lesn. Prom., 1980. 288 p. (In Russian)
2. Li J., Yang B., Chen C., Wu W., Zhang L. Aerial-triangulation aided boresight calibration for a low-cost UAV-lidar system. ISPRS Ann. Photogramm., Remote Sens. Spat. Inf. Sci., 2020, vol. V-1-2020, pp. 245–252. doi: 10.5194/isprs-annals-V-1-2020-245-2020.
3. Li X., Li X., Huang J., Shen Z., Wang B., Yuan Y., Zhang K. Improving PPP–RTK in urban environment by tightly coupled integration of GNSS and INS. J. Geod., 2021, vol. 95, no. 132. doi: 10.1007/s00190-021-01578-6.
4. Ainakulov Zh.Zh., Kuz’min A.G., Mukhamedgaliev A.F., Razakova M.G., Smirnov V.V. Controlling the shutter time of the unmanned aerial vehicle camera. Probl. Inf., 2016, vol. 30, no. 1, pp. 84–92. (In Russian)
5. Zhang W., Cai S., Liang X., Shao J., Hu R., Yu S., Yan G. Cloth simulation-based construction of pit-free canopy height models from airborne LiDAR data. For. Ecosyst., 2020, vol. 7, art. 1, pp. 1–13. doi: 10.1186/s40663-019-0212-0.
6. Titarov P.S. Coordinate accuracy characteristics of CE and LE terrain points. Geoprofi, 2010, no. 1, pp. 52–53. (In Russian)
7. Weinstein B.G., Marconi S., Bohlman S., Zare A., White E. Individual tree-crown detection in RGB imagery using semi-supervised deep learning neural networks. Remote Sens., 2019, vol. 11, no. 11, art. 1309, pp. 1–13. doi: 10.3390/rs11111309.
8. Weinstein B.G., Marconi S., Bohlman S., Zare A., White E. Geographic generalization in airborne RGB deep learning tree detection. bioRxiv, 2019. doi: 10.1101/790071.
9. Gupta S., Sharma P., Sharma D., Gupta V., Sambyal N. Detection and localization of potholes in thermal images using deep neural networks. Multimedia Tools Appl., 2020, vol. 79, pp. 26265–26284. doi: 10.1007/s11042-020-09293-8.
10. Tian H., Zheng Y., Jin Z. Improved RetinaNet model for the application of small target detection in the aerial images. IOP Conf. Ser.: Earth Environ. Sci., 2020, vol. 585, no. 1, art. 012142, pp. 1–8. doi: 10.1088/1755-1315/585/1/012142.

The content is available under the license Creative Commons Attribution 4.0 License.