Skip to main content

Feasibility study and mapping of vegetative landscape structures important for biodiversity and climate change mitigation with remote sensing

Description

The project objective was to map different kinds of vegetation appearances. We used the nationwide aerolaser scanning data ant proposed a method with which we achieved a high level of correct classification (90 %). The method achieved the best results classifying waterside trees, however the poorest were the results classifying of overgrown parcels.

To obtain data with higher temporal resolution we proposed a method that makes use of satellite data. We tested using Sentinel-2 images to map landscape structures and found out there weren’t many false positive results. On the other hand, there were many false negatives. We assume the proposed method would have given us much better results should we use satellite data with higher spatial resolution. The usage of those is still limited due to their high price, though.

primeri_pravilne_klasiifikacije

Examples of positive identification and classification of landscape features. Study sites Barje (a–d), Bled (e–h), Kras (i–l), Vipavska dolina (m–p), Krško (r–t) and Moravske Toplice (u).


Research Project