Találati lista | Tudóstér

001-es BibID:	BIBFORM082766
Első szerző:	Burai Péter (agrármérnök)
Cím:	Airborne hyperspectral remote sensing for identification grassland vegetation / Burai Péter, Tomor Tamás, Bekő László, Deák Balázs
Dátum:	2015
ISSN:	2194-9034
Megjegyzések:	In our study we classified grassland vegetation types of an alkali landscape (Eastern Hungary), using different image classification methods for hyperspectral data. Our aim was to test the applicability of hyperspectral data in this complex system using various image classification methods. To reach the highest classification accuracy, we compared the performance of traditional image classifiers, machine learning algorithm, feature extraction (MNF-transformation) and various sizes of training dataset. Hyperspectral images were acquired by an AISA EAGLE II hyperspectral sensor of 128 contiguous bands (400?1000 nm), a spectral sampling of 5 nm bandwidth and a ground pixel size of 1 m. We used twenty vegetation classes which were compiled based on the characteristic dominant species, canopy height, and total vegetation cover. Image classification was applied to the original and MNF (minimum noise fraction) transformed dataset using various training sample sizes between 10 and 30 pixels. In the case of the original bands, both SVM and RF classifiers provided high accuracy for almost all classes irrespectively of the number of the training pixels. We found that SVM and RF produced the best accuracy with the first nine MNF transformed bands. Our results suggest that in complex open landscapes, application of SVM can be a feasible solution, as this method provides higher accuracies compared to RF and MLC. SVM was not sensitive for the size of the training samples, which makes it an adequate tool for cases when the available number of training pixels are limited for some classes.
Tárgyszavak:	Természettudományok Földtudományok idegen nyelvű folyóiratközlemény külföldi lapban folyóiratcikk Vegetation Mapping Hyperspectral Image Classification Maximum Likelihood Classifier Random Forest Support Vector Machine Open Landscape
Megjelenés:	ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. - XL-3/W3 (2015), p. 427-431. -
További szerzők:	Tomor Tamás (1976-) (geográfus) Bekő László (1986-) (okleveles vidékfejlesztési agrármérnök) Deák Balázs (1978-) (biológus)
Pályázati támogatás:	TÁMOP- 4.2.2.D-15/1/ KONV-2015-0010 TÁMOP
Internet cím:	Szerző által megadott URL DOI Intézményi repozitóriumban (DEA) tárolt változat
Borító:
Saját polcon:

001-es BibID:	BIBFORM081438
Első szerző:	Zlinszky András
Cím:	Biodiversity mapping via Natura 2000 conservation status and ebv assessment using airborne laser scanning in alkali grasslands / A. Zlinszky, B. Deák, A. Kania, A. Schroiff, N. Pfeifer
Dátum:	2016
ISSN:	2194-9034
Megjegyzések:	Biodiversity is an ecological concept, which essentially involves a complex sum of several indicators. One widely accepted such set of indicators is prescribed for habitat conservation status assessment within Natura 2000, a continental-scale conservation programme of the European Union. Essential Biodiversity Variables are a set of indicators designed to be relevant for biodiversity and suitable for global-scale operational monitoring. Here we revisit a study of Natura 2000 conservation status mapping via airbone LIDAR that develops individual remote sensing-derived proxies for every parameter required by the Natura 2000 manual, from the perspective of developing regional-scale Essential Biodiversity Variables. Based on leaf-on and leaf-off point clouds (10 pt/m2) collected in an alkali grassland area, a set of data products were calculated at 0.5x0.5 m resolution. These represent various aspects of radiometric and geometric texture. A Random Forest machine learning classifier was developed to create fuzzy vegetation maps of classes of interest based on these data products. In the next step, either classification results or LIDAR data products were selected as proxies for individual Natura 2000 conservation status variables, and fine-tuned based on field references. These proxies showed adequate performance and were summarized to deliver Natura 2000 conservation status with 80% overall accuracy compared to field references. This study draws attention to the potential of LIDAR for regional-scale Essential Biodiversity variables, and also holds implications for global-scale mapping. These are (i) the use of sensor data products together with habitat-level classification, (ii) the utility of seasonal data, including for non-seasonal variables such as grassland canopy structure, and (iii) the potential of fuzzy mapping-derived class probabilities as proxies for species presence and absence.
Tárgyszavak:	Természettudományok Biológiai tudományok konferenciacikk folyóiratcikk LIDAR Essential Biodiversity Variables Natura 2000 Conservation status Biodiversity assessment
Megjelenés:	International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. - 41 : B8 (2016), p. 1293-1299. -
További szerzők:	Deák Balázs (1978-) (biológus) Kania, Adam (1974-) (program fejlesztő) Schroiff, Anke Pfeifer, Norbert (1971-) (geodéziai mérnök)
Pályázati támogatás:	OTKA PD 115627 OTKA
Internet cím:	Szerző által megadott URL DOI Intézményi repozitóriumban (DEA) tárolt változat
Borító:
Saját polcon: