Paper
17 October 2023 Evaluation of the hyperspectral monitoring method capabilities for forests areas
Michael L. Belov, Aleksey M. Belov, Victor A. Gorodnichev, Sergey V. Alkov, Aleksandr A. Shkarupilo
Author Affiliations +
Proceedings Volume 12780, 29th International Symposium on Atmospheric and Ocean Optics: Atmospheric Physics; 127802Q (2023) https://doi.org/10.1117/12.2690227
Event: XXIX International Symposium "Atmospheric and Ocean Optics, Atmospheric Physics", 2023, Moscow, Russian Federation
Abstract
The purpose of the study is to analyze the possibilities of forest areas hyperspectral monitoring. The mathematical modeling of the forest territories elements classification on the created neural network using experimentally measured reflection coefficients is presented. The simulation results show that hyperspectral monitoring in the spectral range of 400-2400 nm allows for classification of forest elements (different species of deciduous and coniferous trees, deadwood, swamps, water bodies, soils without vegetation, different types of mosses and lichens, post-fire areas, different shrubby plants) with the probability of correct classification of more than 0.73 and the probability of incorrect classification of less than 0.037. The use of additional information from the laser altimeter allows to significantly improve the classification. The created neural network, using hyperspectral monitoring data and lidar data on the height of trees, provides the probabilities of correct forest area elements classification of more than 0.8 and the probabilities of incorrect classification of less than 0.025.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Michael L. Belov, Aleksey M. Belov, Victor A. Gorodnichev, Sergey V. Alkov, and Aleksandr A. Shkarupilo "Evaluation of the hyperspectral monitoring method capabilities for forests areas", Proc. SPIE 12780, 29th International Symposium on Atmospheric and Ocean Optics: Atmospheric Physics, 127802Q (17 October 2023); https://doi.org/10.1117/12.2690227
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Reflection

Mathematical modeling

Neural networks

Vegetation

Hyperspectral simulation

LIDAR

Environmental monitoring

Back to Top