magistrsko delo
Gregor Senegačnik (Author), Milan Kobal (Mentor)

Abstract

V nalogi smo z izvedbo metod daljinskega pridobivanja podatkov in njihove obdelave določali število kosov in volumen lesenega plavja v hudourniških strugah. Izdelan je bil pravi ortofoto posnetek območja, vegetacijski indeks NDVI, oblak točk in rastrski sloj digitalnega modela površja. S pomočjo pridobljenih slojev je bil izračunan volumen plavja s pomočjo štirih različnih metod, ki so temeljile na oblaku točk (minimalni konveksni trirazsežni objekt), rastrskih celicah in dolžinah ter širinah poligonov. Multispektralni posnetki so se izkazali za uporaben pripomoček pri ročnem načinu prepoznavanja lesenega plavja. Slabši rezultati so bili doseženi z metodo avtomatskega določanja lesenega plavja. Metoda je temeljila na klasifikacijah vegetacijskega indeksa NDVI, površin poligonov in klasifikaciji z metodo logistične regresije. Z metodo smo uspešno določili 46 kosov plavja od skupno 144. Iz oblaka točk smo volumen plavja ocenili s 33 % preveliko vrednostjo.

Keywords

leseno plavje;poplavna varnost;fotogrametija;brezpilotni letalniki;vegetacijski indeks NDVI;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL BF - Biotechnical Faculty
Publisher: [G. Senegačnik]
UDC: 630*37+630*38:630*58(043.2)=163.6
COBISS: 5183142 Link will open in a new window
Views: 1346
Downloads: 446
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Assessment of large woody debris in river streams using multispectral images
Secondary abstract: In the master thesis, the methods of remote data sensing and processing determined the number and volume of woody debris in torrential streams. A true orthophoto image of the area, NDVI vegetation index, the point cloud and the raster layer of the digital surface model were created. With the obtained layers and four different methods the volume of woody debris was calculated. The methods were based on a cloud point (minimum convex 3D object), raster cells, lengths and polygon widths. Multispectral imagery has been proven to be a useful tool in recognizing woody debris manually. Inferior results were achieved by the method of automatic determination of woody debris. The method was based on the classification of the NDVI vegetation index, filtering of surfaces and classification of logistic regression. With the method of classification, we successfully determined 46 pieces of woody debris from a total of 144. The best method for volume calculation was the point cloud where woody debris was overestimated by 33 %.
Secondary keywords: woody debris;photogrametry;flood safety;unmanned aerial vehicles;NDVI index;
Type (COBISS): Master's thesis/paper
Study programme: 0
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. Ljubljana, Biotehniška fak., Oddelek za gozdarstvo in obnovljive gozdne vire
Pages: XI f., 64 str., [1] zganj. f. pril.
ID: 10959675