Processing Very High-Resolution Satellite Images for Individual Tree Identification with Local Maxima Method

In the last decades, different Remote Sensing (RS) techniques and instruments were developed and utilized to manage and monitor the natural and semi-natural resources. Increasing the number of the sensors with the high spatial and spectral resolution, the remote sensing techniques and the Geographic Information System (GIS), provide more and detailed information, required for the precision agriculture tasks, and support, where possible, the decision-making process. The aim of this study is to develop a chain process, to obtain by using Earth Observation (EO) data, detailed information about the detection of the olive tree crowns. The Individual Tree Crown (ITC) detection process is implemented in a semi-automatic workflow based on Local Maxima Filter (LMF) applied on the Digital Aerial image and WorldView-2 (WV-2) images. The results indicate that the image data characteristics play a fundamental role to detect trees by EO data. For both datasets, the results show a higher accuracy achieved with the NDVI (Normalized Difference Vegetation Index), highlighting the spectral characteristics of the vegetation in the red and InfraRed domain.