The Science of The Total Environment, Год журнала: 2024, Номер 958, С. 178132 - 178132
Опубликована: Дек. 17, 2024
Язык: Английский
The Science of The Total Environment, Год журнала: 2024, Номер 958, С. 178132 - 178132
Опубликована: Дек. 17, 2024
Язык: Английский
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0Forestry Engineering Journal, Год журнала: 2025, Номер 14(4), С. 63 - 84
Опубликована: Май 7, 2025
The development of Earth remote sensing methods, neural network technologies, creation machine learning models, etc. will allow developing new algorithms, indicators and criteria that significantly increase the efficiency forest monitoring help reduce financial costs. basis for work was verification ultra-high- high-resolution satellite imagery data based on in-situ survey materials conducted permanent test plots in Voronezh region. generated elements training samples using a classifier automated, highly accu-rate determination stand taxation data. When verifying amount aboveground phytomass calculated materials, significant similarity presented results revealed. In 67% cases, average values determined by different methods did not differ from each other (at significance level p<0.05). Reliable differences were found decid-uous stands with high horizontal canopy density, presence second tier abundant understory vegetation. As result work, theoretical laid further research conducting field for-est conservation, protection reproduction. material is useful building multidisciplinary practical areas restoration biological diversity phytocenoses, as well ensuring integrity ecological stability forests, under modern trends carbon cycles climate changes.
Язык: Английский
Процитировано
0The Science of The Total Environment, Год журнала: 2024, Номер 958, С. 178132 - 178132
Опубликована: Дек. 17, 2024
Язык: Английский
Процитировано
1