Detail publikace

Možnosti vylepšení klasifikace obrazu pomocí nástrojů GIS

BARTONĚK, D. BUREŠ, J. OPATŘILOVÁ, I.

Originální název

Possibilities of Improvement of Image Classification via GIS Tools

Český název

Možnosti vylepšení klasifikace obrazu pomocí nástrojů GIS

Anglický název

Possibilities of Improvement of Image Classification via GIS Tools

Typ

článek ve sborníku

Jazyk

en

Originální abstrakt

The article deals with the possibilities of the enhancement of results of image classification using tools of geographic information systems (GIS). Theoretical solutions are based on the theory of rough sets. Spatial analyses in GIS are the implementation tools of the theoretical model. Improved results of image classification are achieved by filtration of added layer with clearly defined classes and also by manually editing in the necessary extent in GIS. The authors of this paper show that the filtration using more than one layer leads to an enormous increase of the complexity of the whole process with inadequate contribution of the quality of classification results. The proposed method was tested in the project of data analysis of storage of gas facilities under certain types of terrain surface in the Czech Republic (CR). This analysis was done in order to determine reproductive values of gas facilities (pipelines) and the valuation of costs which would be necessary to spend for building new networks. The authors solved this project for the GasNet, Ltd. Company which is a part of a RWE group in the Czech Republic. Input data were orthophoto with the resolution of 25 cm/pixel and selected layers of communications of Fundamental Base of Geographic Data of the CR (ZABAGED). Due to the territorial coverage with the area of 64,350 km2, these were massive tasks with data volume of 500 GB. Processing was carried out in ArcGIS 10.0 environment via special created application in Python language with support of the ESRI libraries. The results demonstrated the efficacy (effectiveness) of this process and the reducing of the error rate to 2% - 3 % was achieved over the modeled area for given purpose.

Český abstrakt

Článek se zabývá možnostmi vylepšení výsledků klasifikace obrazu s využitím nástrojů geografických informačních systémů (GIS). Teoretická východiska jsou založena na teorii hrubých množin. Implementačními nástroji teoretického modelu jsou prostorové analýzy v GIS. Vylepšení výsledků klasifikace obrazu je dosaženo jednak filtrací přidané vrstvy s jednoznačně definovanými třídami a jednak ruční editací v nezbytně nutném rozsahu v GIS. Autory tohoto příspěvku bylo prokázáno, že filtrace s využitím více než jedné vrstvy vede k enormnímu zvýšení složitosti celého procesu s neadekvátním přínosem kvality výsledků klasifikace. Navržená metoda byla ověřena v rámci projektu datové analýzy uložení plynárenských zařízení pod určitými typy povrchů terénu na území České republiky za účelem stanovení reprodukčních hodnot plynárenských zařízení (plynovodů) a ocenění nákladů, které by bylo nutné vynaložit na vybudování nových sítí, který řešili autoři pro společnost GasNet, s.r.o., která je součástí koncernu RWE Česká republika (ČR). Vstupními daty bylo ortofoto s rozlišením 25 cm/pixel a vrstvy komunikací ZABAGED ČR. Vzhledem k územnímu pokrytí ČR s rozlohou 64 350 km2 šlo o masivní úlohy s objemem dat 500 GB. Zpracování probíhalo jednak v účelově vytvořené aplikaci v jazyku Python s podporu knihoven ESRI a jednak v prostředí ArcGIS 10.0. Výsledky prokázaly účinnost (efektivnost) tohoto postupu, kterým bylo dosaženo snížení chybovosti až na 2% – 3% na celém modelovaném území.

Anglický abstrakt

The article deals with the possibilities of the enhancement of results of image classification using tools of geographic information systems (GIS). Theoretical solutions are based on the theory of rough sets. Spatial analyses in GIS are the implementation tools of the theoretical model. Improved results of image classification are achieved by filtration of added layer with clearly defined classes and also by manually editing in the necessary extent in GIS. The authors of this paper show that the filtration using more than one layer leads to an enormous increase of the complexity of the whole process with inadequate contribution of the quality of classification results. The proposed method was tested in the project of data analysis of storage of gas facilities under certain types of terrain surface in the Czech Republic (CR). This analysis was done in order to determine reproductive values of gas facilities (pipelines) and the valuation of costs which would be necessary to spend for building new networks. The authors solved this project for the GasNet, Ltd. Company which is a part of a RWE group in the Czech Republic. Input data were orthophoto with the resolution of 25 cm/pixel and selected layers of communications of Fundamental Base of Geographic Data of the CR (ZABAGED). Due to the territorial coverage with the area of 64,350 km2, these were massive tasks with data volume of 500 GB. Processing was carried out in ArcGIS 10.0 environment via special created application in Python language with support of the ESRI libraries. The results demonstrated the efficacy (effectiveness) of this process and the reducing of the error rate to 2% - 3 % was achieved over the modeled area for given purpose.

Klíčová slova

Klasifikace obrazu, GIS, prostorová analýza

Rok RIV

2014

Vydáno

15.06.2014

Nakladatel

Bulgarian Cartographic Assosiation

Místo

Sofia, Bulgaria

ISBN

0-470-85151-1

Kniha

International Conference on Cartography and GIS

Číslo edice

1

Strany od

96

Strany do

102

Strany počet

7

BibTex


@inproceedings{BUT108231,
  author="Dalibor {Bartoněk} and Jiří {Bureš} and Irena {Opatřilová}",
  title="Possibilities of Improvement of Image Classification via GIS Tools",
  annote="The article deals with the possibilities of the enhancement of results of image classification using tools of geographic information systems (GIS). Theoretical solutions are based on the theory of rough sets. Spatial analyses in GIS are the implementation tools of the theoretical model. Improved results of image classification are achieved by filtration of added layer with clearly defined classes and also by manually editing in the necessary extent in GIS. The authors of this paper show that the filtration using more than one layer leads to an enormous increase of the complexity of the whole process with inadequate contribution of the quality of classification results. The proposed method was tested in the project of data analysis of storage of gas facilities under certain types of terrain surface in the Czech Republic (CR). This analysis was done in order to determine reproductive values of gas facilities (pipelines) and the valuation of costs which would be necessary to spend for building new networks. The authors solved this project for the GasNet, Ltd. Company which is a part of a RWE group in the Czech Republic. Input data were orthophoto with the resolution of 25 cm/pixel and selected layers of communications of Fundamental Base of Geographic Data of the CR (ZABAGED). Due to the territorial coverage with the area of 64,350 km2, these were massive tasks with data volume of 500 GB. Processing was carried out in ArcGIS 10.0 environment via special created application in Python language with support of the ESRI libraries. The results demonstrated the efficacy (effectiveness) of this process and the reducing of the error rate to 2% - 3 % was achieved over the modeled area for given purpose.",
  address="Bulgarian Cartographic Assosiation",
  booktitle="International Conference on Cartography and GIS",
  chapter="108231",
  howpublished="electronic, physical medium",
  institution="Bulgarian Cartographic Assosiation",
  number="1",
  year="2014",
  month="june",
  pages="96--102",
  publisher="Bulgarian Cartographic Assosiation",
  type="conference paper"
}