Uporaba lidarskih podatkov za klasifikacijo pokrovnosti

Abstract

Keywords

Data

Language:	Slovenian
Year of publishing:	2009
Source:	Ljubljana
Typology:	2.09 - Master's Thesis
Organization:	UL FGG - Faculty of Civil and Geodetic Engineering
Publisher:	[A. Švab Lenarčič]
UDC:	528.8:711.14(043.3)
COBISS:	4543329
Views:	2007
Downloads:	647
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Use of lidar data for land cover classification
Secondary abstract:	While the majority of land cover maps of Slovenia is intended to serve analyses at the level of country or statistical region, the thesis discusses how to create a map that could serve local projects as well, by processing various lidar data, such as elevation, intensity and standard deviation. Observations of several researchers have been confirmed: the intensity of the observed object depends on many factors and is therefore not easy to interpret. The categories of grass, field and asphalt could nevertheless be classified distinctively by carefully studying the problem and implementing various improvements. The use of intensity alone, however, could not extract trees and buildings, so standard deviation of hight was calculated from the lidar data, which has actually proved very successful. On the basis of independent lidar data, a three dimensional high quality land cover map of local area has been successfully generated, which distinguishes five basic categories, although more detailed sub-categories could be introduced as well, if necessary. Once the methodology is determined, products can be generated fully automatically, with minimal effort and costs. What is even more important, the resulting maps are of very high position and thematic accuracy. Advantages of such maps are their usefulness for local studies, prompt and cost-friendly method of layer acquisition, time adjustment of data, objective land cover identification, GIS use possibility, possibility of layer adjustment for specific studies, possibility of frequent layer updating, and last but not least, the possibility of three dimensional map display. There is also one main disadvantage: the map is useful merely for the study on a local level, since the huge quantity of point data requires very powerful hardware and software technology.
Secondary keywords:	geodesy;master of science thesis;land cover;lidar;intensity;
URN:	URN:NBN:SI
File type:	application/pdf
Type (COBISS):	Master's thesis
Thesis comment:	Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo
Pages:	XXIV, 135 str.
Type (ePrints):	thesis
Title (ePrints):	Use of lidar data for land cover classification
Keywords (ePrints):	pokrovnost;lidar;intenziteta
Keywords (ePrints, secondary language):	land cover;lidar;intensity
Abstract (ePrints):	Ker večina kart pokrovnosti v Sloveniji zadošča zgolj analizam na ravni države oz. statistične regije, sem v nalogi skušala izdelati tudi karto za uporabo na lokalni ravni. Pri tem sem kot vir uporabila različne podatke lidarskega snemanja, kot so višina, intenziteta in standardna deviacija višine. Potrdila sem dejstvo številnih raziskovalcev, da je intenziteta opazovanega objekta odvisna od vrste dejavnikov in zato precej težka za interpretacijo. Po njeni preučitvi in nekaterih izboljšavah sem z njo kljub vsemu zelo dobro izločila kategorije trava, njive in asfalt. Za izločitev dreves in stavb uporaba intenzitete žal ni nudila ustreznih rezultatov, zato sem iz lidarskih podatkov izračunala standardno deviacijo višin, ki se je za izločitev teh dveh kategorij izkazala kot zelo uspešna. V sklopu naloge sem na osnovi samostojnih lidarskih podatkov uspela izdelati kakovostno karto pokrovnosti lokalnega območja, ki loči pet osnovnih kategorij, po potrebi pa bi jih lahko ločili tudi več. Ko je metodologija izdelave karte znana, lahko karto izdelamo povsem avtomatsko, z minimalnim naporom in stroški. Pri tem dobimo karto zelo dobre položajne in tematske natančnosti. Prednosti takšnih kart so uporabnost za lokalne študije, hiter in cenovno ugoden način pridobitve sloja, časovna usklajenost podatkov, objektivna določitev pokrovnosti, možnost uporabe v GIS, možnost prilagoditve sloja za specifične študije, možnost pogostega osveževanja sloja ter možnost prikaza karte v trirazsežnem prostoru. Slabost je, da je karta uporabna zgolj za študije na lokalni ravni, saj velika količina točkovnih podatkov zahteva zahtevno strojno in programsko opremo.
Abstract (ePrints, secondary language):	While the majority of land cover maps of Slovenia is intended to serve analyses at the level of country or statistical region, the thesis discusses how to create a map that could serve local projects as well, by processing various lidar data, such as elevation, intensity and standard deviation. Observations of several researchers have been confirmed: the intensity of the observed object depends on many factors and is therefore not easy to interpret. The categories of grass, field and asphalt could nevertheless be classified distinctively by carefully studying the problem and implementing various improvements. The use of intensity alone, however, could not extract trees and buildings, so standard deviation of hight was calculated from the lidar data, which has actually proved very successful. On the basis of independent lidar data, a three dimensional high quality land cover map of local area has been successfully generated, which distinguishes five basic categories, although more detailed sub-categories could be introduced as well, if necessary. Once the methodology is determined, products can be generated fully automatically, with minimal effort and costs. What is even more important, the resulting maps are of very high position and thematic accuracy. Advantages of such maps are their usefulness for local studies, prompt and cost-friendly method of layer acquisition, time adjustment of data, objective land cover identification, GIS use possibility, possibility of layer adjustment for specific studies, possibility of frequent layer updating, and last but not least, the possibility of three dimensional map display. There is also one main disadvantage: the map is useful merely for the study on a local level, since the huge quantity of point data requires very powerful hardware and software technology.
Keywords (ePrints, secondary language):	land cover;lidar;intensity
ID:	8311017