Charles Darwin University

CDU eSpace
Institutional Repository

 
CDU Staff and Student only
 

Automating image classification, a preliminary foray using archival GIS data to label pixels

Fegan, Matthew Terence, Devonport, Christopher and Ahmad, Waqar (2001). Automating image classification, a preliminary foray using archival GIS data to label pixels. In: 2001 International Geoscience and Remote Sensing Symposium (IGARSS 2001), Sydney, NSW, 9-13 July 2001.

Document type: Conference Paper
Citation counts: Google Scholar Search Google Scholar

Author Fegan, Matthew Terence
Devonport, Christopher
Ahmad, Waqar
Title Automating image classification, a preliminary foray using archival GIS data to label pixels
Conference Name 2001 International Geoscience and Remote Sensing Symposium (IGARSS 2001)
Conference Location Sydney, NSW
Conference Dates 9-13 July 2001
Conference Publication Title Scanning the Present and Resolving the Future: Proceedings of the IEEE 2001 International Geoscience and Remote Sensing Symposium
Place of Publication United States of America
Publisher IEEE Conference Publications
Publication Year 2001
Volume Number 4
ISBN 0-7803-7031-7   (check CDU catalogue  open catalogue search in new window)
Start Page 1886
End Page 1888
Total Pages 3
Abstract This paper proposes a procedure to update existing mapping from a classified image in the Mary River floodplain of the Northern Territory, Australia. This procedure explicitly rests on the assumption that the spatial distribution of land cover types depicted in the reference cover type mapping is "still mostly correct", (and contains sufficient still-valid information to train a classification of a more recent image), with a view to updating the cover type mapping. Empirical probabilities of spectral class to cover type were estimated from GIS overlay of a classified TM image over existing mapping. Cover types are additionally characterised as mosaics of characteristic (template) textures. The image texture of a window around each pixel was estimated and compared for similarity to the GIS derived cover type texture templates. Pixels were labelled to the most 'likely' cover type. The 'likelihood' of a given cover type label was estimated by combining both the similarity of pixel neighbourhood texture and the the pixel spectral class probability of association to that cover type. The resulting map does not conform completely to the original mapping; however enough cover types do map to plausible locations to warrant continuing the investigation.
Description for Link Link to published version
URL http://ieeexplore.ieee.org/document/977105/
 
Versions
Version Filter Type
Access Statistics: 5 Abstract Views  -  Detailed Statistics
Created: Fri, 29 Aug 2014, 19:42:57 CST by Anthony Hornby