Application of remote sensing data for monitoring of forest vegetation on the territory of nature park “Blue Stones," Bulgaria

2019 ◽  
Author(s):  
Andrey Stoyanov ◽  
Nikolay Georgiev ◽  
Iliyana Gigova ◽  
Denitsa Borisova
Keyword(s):  
Remote Sensing ◽  
Remote Sensing Data ◽  
Forest Vegetation ◽  
Sensing Data ◽  
Nature Park

scholarly journals Indicative landscape monitoring of the national nature park (case study the territory of Slobozhansky National Nature Park)

2018 ◽  
Keyword(s):  
Remote Sensing ◽  
Remote Sensing Data ◽  
Terrestrial Ecosystems ◽  
Gis Technology ◽  
Landscape Indicators ◽  
Monitoring Method ◽  
Sensing Data ◽  
The Usa ◽  
Nature Park ◽  
Landscape Monitoring

Formulation of the problem. National Natural Parks (NNP) – protected areas where anthropogenic and natural landscapes are combined in the same territory. In addition, the main functions of such objects are significantly competitive, which requires monitoring of changes in existing landscapes. It is necessary to define the local objects which, being the most sensitive, at the same time have small plasticity, therefore, are capable to react quickly and adequately to any changes. That is what we call indicative. Analysis of recent research and publications. Many researchers of the USA, Great Britain, Germany, Australia conduct landscape monitoring using remote sensing data and GIS technologies. For example, D. Keith, S. Rodoreda, L. Holman, R. Noss, U. Walz, and others. The National Inventory of Landscapes in Sweden studies development of modern landscape monitoring in countries of Europe. Landscape Monitoring of Terrestrial Ecosystems, studied by researches R. Kennedy, J. Jons, K. Jones and others allow using data of satellite for selection of plant contours using Gis-technology. Landscape monitoring of the territory of NNP «Slobozhanskiy» has never been carried out. The aim of the study is to choose satellite images, taking into account the area of the study, the choice of optimal methods of their processing for the compilation of a database of landscape structure facies for landscape monitoring based on long-term observations on the ground, comparing their results with geodata. We have determined wetlands, as landscape indicators. Presentation of the main material of the study. Comprehensive analysis of remote sensing data carried out by the authors, allowed us to make sure that vegetation cover is the most indicative, except for the contours of wetlands, which are clearly identified and easily compared in multi-spectral images. It is reliably determined by the characteristic features combine with the corresponding spectral ranges and the image structure. In addition, changes in vegetation allows you to visually determine changes in landscape groupings and the speed of these changes. Summary. The indicative features of landscape monitoring are wetlands, and there are two direct indicators: the contours of wetlands and the change in the aspect of vegetation. The monitoring method is a multispectral analysis of images obtained by processing combinations of spectral channels, which showed the ability to determine the changes in the selection, taking into account reflectivity of the surface. Limitations of the method are the following: there is no established method of meticulous analysis of changes in the structure of vegetation, which is observed visually, but is not reflected instrumentally; inability to take into account random features of the territory conditions and space scanning at a certain point, which is interesting for the study. Finally, the types of monitoring objects, indicative signs of changes and ways to track them according to high-precision and generally available satellite information are determined.

scholarly journals The t-SNE Algorithm as a Tool to Improve the Quality of Reference Data Used in Accurate Mapping of Heterogeneous Non-Forest Vegetation

2019 ◽  
Vol 12 (1) ◽  
pp. 39 ◽  
Author(s):  
Anna Halladin-Dąbrowska ◽  
Adam Kania ◽  
Dominik Kopeć
Keyword(s):  
Remote Sensing ◽  
Reference Data ◽  
Remote Sensing Data ◽  
Ground Truth ◽  
Forest Vegetation ◽  
Visual Interpretation ◽  
Reference Dataset ◽  
Visual Evaluation ◽  
Sensing Data

Supervised classification methods, used for many applications, including vegetation mapping require accurate “ground truth” to be effective. Nevertheless, it is common for the quality of this data to be poorly verified prior to it being used for the training and validation of classification models. The fact that noisy or erroneous parts of the reference dataset are not removed is usually explained by the relatively high resistance of some algorithms to errors. The objective of this study was to demonstrate the rationale for cleaning the reference dataset used for the classification of heterogeneous non-forest vegetation, and to present a workflow based on the t-distributed stochastic neighbor embedding (t-SNE) algorithm for the better integration of reference data with remote sensing data in order to improve outcomes. The proposed analysis is a new application of the t-SNE algorithm. The effectiveness of this workflow was tested by classifying three heterogeneous non-forest Natura 2000 habitats: Molinia meadows (Molinion caeruleae; code 6410), species-rich Nardus grassland (code 6230) and dry heaths (code 4030), employing two commonly used algorithms: random forest (RF) and AdaBoost (AB), which, according to the literature, differ in their resistance to errors in reference datasets. Polygons collected in the field (on-ground reference data) in 2016 and 2017, containing no intentional errors, were used as the on-ground reference dataset. The remote sensing data used in the classification were obtained in 2017 during the peak growing season by a HySpex sensor consisting of two imaging spectrometers covering spectral ranges of 0.4–0.9 μm (VNIR-1800) and 0.9–2.5 μm (SWIR-384). The on-ground reference dataset was gradually cleaned by verifying candidate polygons selected by visual interpretation of t-SNE plots. Around 40–50% of candidate polygons were ultimately found to contain errors. Altogether, 15% of reference polygons were removed. As a result, the quality of the final map, as assessed by the Kappa and F1 accuracy measures as well as by visual evaluation, was significantly improved. The global map accuracy increased by about 6% (in Kappa coefficient), relative to the baseline classification obtained using random removal of the same number of reference polygons.

Thematic services for analysis of forests using earth observation data

2020 ◽  
Vol 5 ◽  
pp. 175-181
Author(s):  
V.A. Zelentsov ◽  
◽  
M.R. Ponomarenko ◽  
I.Y. Pimanov
Keyword(s):  
Remote Sensing ◽  
Remote Sensing Data ◽  
Earth Observation ◽  
Forest Vegetation ◽  
Observation Data ◽  
Earth Remote Sensing ◽  
Sensing Data ◽  
Earth Observation Data

The paper presents an overview of existing thematic services based on Earth remote sensing data from space and aimed at monitoring and analysis of forest vegetation and dynamics of its changes.

Assessment of the national market demand for high-resolution remote sensing data

2002 ◽  
Vol 8 (1) ◽  
pp. 15-22
Author(s):  
V.N. Astapenko ◽  
◽  
Ye.I. Bushuev ◽  
V.P. Zubko ◽  
V.I. Ivanov ◽  
...  
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Remote Sensing Data ◽  
Market Demand ◽  
Sensing Data ◽  
National Market
Sign in / Sign up

Export Citation Format

Share Document