Publication detail

Visualization of Uncertainty in LANDSAT Classification Process

ŠŤASTNÝ, J. ŠKORPIL, V. FEJFAR, J.

Original Title

Visualization of Uncertainty in LANDSAT Classification Process

Czech Title

Vizualizace neurčitosti v LANDSAT klasifikačním procesu

English Title

Visualization of Uncertainty in LANDSAT Classification Process

Type

conference paper

Language

en

Original Abstract

Many uncertainties can be found in the classification of remotely sensed data. Namely they can arise in defining classification classes. We use two ways, incorporating acquired Corine Land Cover labels and our manually annotated labels. We are describing several visualization possibilities to demonstrate uncertainties in labels and their connections with classification results. We use parallel coordinates to visualize data, presenting problems in classes definitions. These failures can be consequently seen in the results of classification in confusion matrix.

Czech abstract

Při klasifikace dálkově snímaných dat je možno nalézt množství neurčitostí. Vzniknout mohou zejména při definování klasifikačních tříd. Používáme dva způsoby, a to zařazování získaných Corine Land Cover návěští a naše ručně vytvořená návěští. Popisujeme několik možností vizualizace demonstrující neurčitosti návěští a jejich spojení s výsledky klasifikace. Pro vizualizaci dat používáme paralelní přiřazení a ukazujeme problémy při definování tříd. Tyto neurčitosti mohou být následně vidět ve výsledcích klasifikace v příslušné matici.

English abstract

Many uncertainties can be found in the classification of remotely sensed data. Namely they can arise in defining classification classes. We use two ways, incorporating acquired Corine Land Cover labels and our manually annotated labels. We are describing several visualization possibilities to demonstrate uncertainties in labels and their connections with classification results. We use parallel coordinates to visualize data, presenting problems in classes definitions. These failures can be consequently seen in the results of classification in confusion matrix.

Keywords

LANDSAT, Corine Land Cover, parallel coordinates, k-NN, Python

RIV year

2014

Released

01.07.2014

Publisher

Brno University of Technology

Location

Berlin

ISBN

978-1-4799-8497-8

Book

Proceedings of the 38th International conference on Telecommunications and Signal Processing

Edition number

1

Pages from

789

Pages to

792

Pages count

4

BibTex


@inproceedings{BUT110027,
  author="Jiří {Šťastný} and Vladislav {Škorpil} and Jiří {Fejfar}",
  title="Visualization of Uncertainty in LANDSAT Classification Process",
  annote="Many uncertainties can be found in the classification of remotely sensed data. Namely they can arise in defining classification classes. We use two ways, incorporating acquired Corine Land Cover labels and our manually annotated labels. We are describing several visualization possibilities to demonstrate uncertainties in labels and their connections with classification results. We use parallel coordinates to visualize data, presenting problems in classes definitions. These failures can be consequently seen in the results of classification in confusion matrix.",
  address="Brno University of Technology",
  booktitle="Proceedings of the 38th International conference on Telecommunications and Signal Processing",
  chapter="110027",
  howpublished="print",
  institution="Brno University of Technology",
  year="2014",
  month="july",
  pages="789--792",
  publisher="Brno University of Technology",
  type="conference paper"
}