<![CDATA[Vegetation is a fundamental component of ecosystems that maintains carbon levels, hydrological cycles, mitigating greenhouse gases, and ensures climate stability. In recent years, the impacts of global climate change have led to changes in vegetation cover at various levels. Efforts to monitor changes in vegetation are important and beneficial for various fields such as forest monitoring, agriculture, and plantations, among others. The main objective of this research is to detect changes both increase and decrease in vegetation using multi-sensor hyperspectral imagery. The hyperspectral images used in this study are Hyperion 2014 and PRISMA 2021. The method involves creating different levels of spectral resolution simulations from hyperspectral images to detect vegetation changes. Meanwhile, the vegetation change Clustering method employs unsupervised (k-means) techniques. The cluster results can indicate vegetation changes such as vegetation degradation, vegetation, devegetation, or no change, though they currently have low accuracy. The highest accuracy is by Simulated RapidEye image simulations, is 33.5%. The low accuracy results attributed insufficient preprocessing, particularly in topographic correction. Additionally, this research indicates that the spectral resolution levels do not have a significant impact on vegetation change detection, as the differences in change classes at each level are very small.]]>
Copyrights © 2024
