Wyniki wyszukiwania - BazTech

Ograniczanie wyników

1 Annals of Computer Science and Information Systems

1 2024

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: d’Alembert operator

Sortuj według:

Ogranicz wyniki do:

d'Alembert Convolution for Enhanced Spatio-Temporal Analysis of Forest Ecosystems

Maskeliūnas Rytis, Damaševičius Robertas

Annals of Computer Science and Information Systems

2024

Vol. 39

141--148

This paper presents a novel approach to enhance the spatiotemporal analysis of forest ecosystems using the d'Alembert convolution method. As forests face increasing threats from climate change, deforestation, and other environmental stresses, there is a growing need for advanced methods that provide more accurate and dynamic information on forest health and management. The d'Alembert convolution, integrating elements from wave equation theory and convolutional neural networks, enables the comprehensive analysis of remote sensing images by capturing both spatial and temporal variations. This methodology not only improves feature extraction, but also helps address the challenges associated with traditional image processing techniques, which often overlook the temporal dynamics of forests. Our results from applying the d'Alembert convolution to two benchmark datasets---DeepGlobe Land Cover 2018 and LoveDA---demonstrate significant improvements in the analysis of forest ecosystems. Specifically, the d'Alembert Network achieved superior performance metrics compared to existing methods, including higher accuracy in classifying various forest types and more effective monitoring of changes over time. This study confirms the potential of the d'Alembert convolution in enhancing forest ecosystem analysis, contributing to smarter forestry practices and better environmental stewardship.