Analysis of Capabilities of the Multispectral Optical Method in Monitoring the Forest Territories

Abstract

The paper analyzes possibilities of the multispectral remote optical method in monitoring the forest areas. Results of mathematical simulation are provided of classification of the forest areas elements in the created neural network using experimentally measured reflection of the forest vegetation coefficients. It is demonstrated that the created neural network ensures high probability of correct classification within the classification problem (according to the multispectral remote optical monitoring data) of the forest probed areas. The selected spectral probing channels in a wide spectral range of ~ 400--2400 nm and the created neural network used seven spectral channels in the visible and in the near infrared spectral range, as well as the active laser sensor to measure the trees height. They provided a probability of correct classification of the forest areas elements (green deciduous trees, green coniferous trees, dry deciduous and coniferous trees, swamps, pastures with different vegetation cover and different types of soils) of more than 0.74 and the probability of misclassification of the forest areas elements of less than 0.08. The multispectral remote optical method could be used in operational monitoring of the vast forest areas from an aircraft (light aircraft or unmanned aerial vehicle)

Please cite this article in English as:

Belov M.L., Belov A.M., Gorodnichev V.A., et al. Analysis of capabilities of the multispectral optical method in monitoring the forest territories. Herald of the Bauman Moscow State Technical University, Series Instrument Engineering, 2022, no. 4 (141), pp. 56--69 (in Russ.). DOI: https://doi.org/10.18698/0236-3933-2022-4-56-69

References

[1] White J.C., Coops N.C., Wulder M.A., et al. Remote sensing technologies for enhancing forest inventories: a review. Can. J. Remote Sens., 2016, vol. 42, no. 5, pp. 619--641. DOI: https://doi.org/10.1080/07038992.2016.1207484

[2] Immitzer M., Vuolo F., Atzberger C. First experience with Sentinel-2 data for crop and tree species classifications in central Europe. Remote Sens., 2016, vol. 8, no. 3, art. 166. DOI: https://doi.org/10.3390/rs8030166

[3] Michez A., Piegay H., Lisein J., et al. Classification of riparian forest species and health condition using multi-temporal and hyperspatial imagery from unmanned aerial system. Environ. Monit. Assess., 2016, vol. 188, no. 3, art. 146. DOI: https://doi.org/10.1007/s10661-015-4996-2

[4] Holzwarth S., Thonfeld F., Abdullahi S., et al. Earth observation based monitoring of forests in Germany: a review. Remote Sens., 2020, vol. 12, no. 21, art. 3570. DOI: https://doi.org/10.3390/rs12213570

[5] Lister A.J., Andersen H., Frescino T., et al. Use of remote sensing data to improve the efficiency of national forest inventories: a case study from the United States national forest inventory. Forests, 2020, vol. 11, no. 12, art. 1364. DOI: https://doi.org/10.3390/f11121364

[6] John E., Bunting P., Hardy A., et al. Forest monitoring system for Tanzania. Remote Sens., 2021, vol. 13, no. 16, art. 3081. DOI: https://doi.org/10.3390/rs13163081

[7] Biswas S., Huang Q., Anand A., et al. A multi sensor approach to forest type mapping for advancing monitoring of sustainable development goals (SDG) in Myanmar. Remote Sens., 2020, vol. 12, no. 19, art. 3220. DOI: https://doi.org/10.3390/rs12193220

[8] Miranda E., Mutiara A.B. Forest classification method based on convolutional neural networks and Sentinel-2 satellite imagery. Int. J. Fuzzy Log. Intell. Syst., 2019, vol. 19, no. 4, pp. 272--282. DOI: https://doi.org/10.5391/IJFIS.2019.19.4.272

[9] Egorov V.D., Kozoderov V.V. Recognition of forest vegetation by aircraft hyper-spectral data. Issledovanie Zemli iz kosmosa, 2016, no. 3, pp. 47--58 (in Russ.).

[10] Kozoderov V.V., Egorov V.D. Pattern recognition of forest canopy using the airborne hyperspectral data and multi-bands high spatial resolution satellite sensor WorldView-2 data. A results comparison and accuracy estimation. Issledovanie Zemli iz kosmosa, 2019, no. 6, pp. 89--102 (in Russ.). DOI: https://doi.org/10.31857/S0205-96142019689-102

[11] Karpov A.A., Bogdanov A.P., Pirtskhalava-Karpova N.R., et al. Using remote sensing for monitoring of reforestation in boreal forest. Izvestiya Sankt-Peterburgskoy lesotekhnicheskoy akademii, 2019, no. 229, pp. 23--43 (in Russ.). DOI: https://doi.org/10.21266/2079-4304.2019.229.23-43

[12] Lebedeva T.A., Kopylova Yu.Yu., Gagarin A.I. Complex monitoring and ecological and economical assessment of forest lands in the territories of intensive subsurface use. Izvestiya Samarskogo nauchnogo tsentra RAN [Izvestia RAS SamSC], 2015, vol. 17, no. 5 (2), pp. 500--504 (in Russ.).

[13] Monitoring lesnykh territoriy [Monitoring of forest territories]. zaomkt.ru: website (in Russ.). Available at: http://zaomkt.ru/monitoring-lesnyh-territoriy (accessed: 02.05.2022).

[14] Monitoring lesnykh khozyaystv [Forestry monitoring]. innoter.com: website (in Russ.). Available at: https://innoter.com/otraslevye-resheniya/lesnoe-khozyaystvo/monitoring-lesnykh-khozyaystv (accessed: 02.05.2022).

[15] Aerokosmicheskiy monitoring lesnogo khozyaystva [Aerospace forestry monitoring]. sovzond.ru: website (in Russ.). Available at: https://sovzond.ru/industry-solutions/forestry (accessed: 02.05.2022).

[16] BPLA dlya lesnogo khozyaystva [Unmanned aircraft for forestry]. geosalut.ru: website (in Russ.). Available at: https://www.geosalut.ru/bpla/monitoring-lesnyh-ugodij/monitoring-lesnyh-ugodij-s-primeneniem-bpla (accessed: 02.05.2022).

[17] Meerdink S.K., Hook S.J., Abbott E.A., et al. ECOSTRESS spectral library --- Version 1.0. 2018. Available at: https://speclib.jpl.nasa.gov (accessed: 02.05.2022).

[18] Clark R.N., Swayze G.A., Wise R., et al. USGS Digital Spectral Library splib06a. U.S. Geological survey, data series 231, 2007. Available at: http://speclab.cr.usgs.gov/spectral.lib06 (accessed: 02.05.2022).

[19] Brunton S.L., Kutz J.N. Neural networks and deep learning. Cambridge, Cambridge University Press, 2019.

[20] Haykin S.S. Neural networks and learning machines. Upper Saddle River, Pearson, 2009.

[21] Duchi J., Hazan E., Singer Y. Adaptive subgradient methods for online learning and stochastic optimization. J. Mach. Learn. Res., 2011, vol. 12, pp. 2121--2159.

[22] Metody optimizatsii neyronnykh setey [Optimization methods for neural networks]. habr.com: website (in Russ.). Available at: https://habr.com/ru/post/318970/ (accessed: 02.05.2022).

[23] Kashirina I.L., Demchenko M.V. Research and comparative analysis of optimization methods used in the teaching of neural networks. Vestnik VGU, seriya: Sistemnyy analiz i informatsionnye tekhnologii [Proceedings of Voronezh State University. Series: Systems Analysis and Information Technologies], 2018, no. 4, pp. 123--132 (in Russ.).

[24] Belov M.L., Belov A.M., Gorodnichev V.A., et al. Laser airborne reflection method for remote sensing forest species composition. IOP Conf. Ser.: Mater. Sc. Eng., 2020, vol. 862, art. 022003. DOI: https://doi.org/10.1088/1757-899X/862/2/022003

[25] Belov M.L., Belov A.M., Gorodnichev V.A., et al. Analysing detection bands of two-spectral reflection method to identify forest species composition. J. Phys.: Conf. Ser., 2021, vol. 2094, art. 042035. DOI: https://doi.org/10.1088/1742-6596/2094/4/042035

[26] Belov M.L., Belov A.M., Gorodnichev V.A., et al. Analysing laser reflection method capabilities to monitor forestland condition and species composition. Proc. SPIE, 2021, vol. 11916. DOI: https://doi.org/10.1117/12.2602893

[27] Wu W., Zhang Z., Zheng L., et al. Research progress on the early monitoring of pine wilt disease using hyperspectral techniques. Sensors, 2020, vol. 20, no. 13, art. 3729. DOI: https://doi.org/10.3390/s20133729

[28] Hyperion. usna.edu: website. Available at: https://www.usna.edu/ users/oceano/pguth/md_help/html/hyperion.htm (accessed: 02.05.2022).

[29] Resurs-P. russianspacesystems.ru: website (in Russ.). Available at: http://russianspacesystems.ru/bussines/dzz/resurs-p (accessed: 02.05.2022).

[30] Hu Y., Wu F., Sun Z., et al. The laser vegetation detecting sensor: a full waveform, large-footprint, airborne laser altimeter for monitoring forest resources. Sensors, 2019, vol. 19, no. 7, art. 1699. DOI: https://doi.org/10.3390/s19071699

[31] d’Oliveira M.V.N., Broadbent E.N., Oliveira L.C., et al. Aboveground biomass estimation in Amazonian tropical forests: a comparison of aircraft- and GatorEye UAV-borne LiDAR data in the Chico Mendes extractive reserve in Acre, Brazil. Remote Sens., 2020, vol. 12, no. 11, art. 1754. DOI: https://doi.org/10.3390/rs12111754

[32] M’ayr’a J., Keski-Saari S., Kivinen S., et al. Tree species classification from airborne hyperspectral and LiDAR data using 3D convolutional neural networks. Remote Sens. Environ., 2021, vol. 256, art. 112322. DOI: https://doi.org/10.1016/j.rse.2021.112322