Use of the Bradley–Terry model to assess uncertainty in an error matrix from a hierarchical segmentation of an ASTER image

A. Stein, Gerrit Gort, Arko Lucieer

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

Abstract

Remotely sensed images are increasingly being used for collection of spatial information. A wide development in sensor systems has occurred during the last decades, resulting in improved spatial, temporal, and spectral resolution. The collection of data by remote sensing is generally more efficient and cheaper than by direct observation and measurement on the ground, although still of a varying quality. Data collected by sensors may be affected by atmospheric factors between sensors and the values reflected on the earth’s surface, local impurities on the earth’s surface, technical deficiencies of sensors and other factors. In addition, only the reflection of the sensor’s signal or of the sunlight on the earth’s surface is being measured, and no direct measurements are made. Consequently, the quality of maps produced by remote sensing needs to be assessed.

Original languageEnglish
Title of host publicationSignal and Image Processing for Remote Sensing
EditorsC.H. Chen
Place of PublicationBoca Raton
PublisherCRC Press/Balkema
Chapter26
Pages591-605
Number of pages15
ISBN (Electronic)9781420003130
ISBN (Print)978-0-8493-5091-7
Publication statusPublished - 2007

    Fingerprint

Keywords

  • EOS
  • ADLIB-ART-247

Cite this

Stein, A., Gort, G., & Lucieer, A. (2007). Use of the Bradley–Terry model to assess uncertainty in an error matrix from a hierarchical segmentation of an ASTER image. In C. H. Chen (Ed.), Signal and Image Processing for Remote Sensing (pp. 591-605). Boca Raton: CRC Press/Balkema.