The use of satellite imagery and Remote Sensing (RS) techniques for vegetation monitoring is increasingly recognized and used in several research fields. Multi-temporal analysis of multispectral images and derived vegetation indices are becoming a useful tool to detect forestry vegetation types. However, the use of multi-temporal images is a time-consuming activity for searching and downloading data and need a huge amount of storage space and workstations with high computing performance. To solve the problem, in this paper, it is proposed the use of Google Earth Engine (GEE) as cloud platform for geospatial analysis, which allows visualization and processing of RS. The paper is articulated in three sections and presents the first results of an ongoing research, on the use of GEE to analyze Sentinel-2 images and derived vegetation indices time-series to compute a supervised classification in order to map forestry vegetation types. In the first section an introduction about RS and supervised and unsupervised classification methods to map forestry vegetation is given. In the second section the proposed method is explained and tested in a square plot of 100 km2 inside the protected area of Aspromonte National Park, in the South of Italy. In the last one, the main results of classification process are analyzed.
CITATION STYLE
Praticò, S., Di Fazio, S., & Modica, G. (2021). Multi temporal analysis of sentinel-2 imagery for mapping forestry vegetation types: A google earth engine approach. In Smart Innovation, Systems and Technologies (Vol. 178 SIST, pp. 1650–1659). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-48279-4_155
Mendeley helps you to discover research relevant for your work.