• DocumentCode
    144385
  • Title

    Canadian arctic shoreline mapping using RADARSAT-2 and optical data through object-based classification

  • Author

    Demers, Anne-Marie ; Banks, Sarah N. ; Pasher, Jon ; Duffe, Jason

  • Author_Institution
    Nat. Environ. Emergencies Centre, Environ. Canada, Montreal, QC, Canada
  • fYear
    2014
  • fDate
    13-18 July 2014
  • Firstpage
    5134
  • Lastpage
    5136
  • Abstract
    Environment Canada has an important role in preparing for environmental emergencies along the coast. Despite this, there is a significant information gap regarding the shoreline types found in the Canadian Arctic, compared to more southern regions. As such, the focus of this analysis was to assess the potential for accurate classification of a number of shoreline types using an object-based approach. Two study areas were assessed, including: Richards Island, Northwest Territories and Ivvavik, Yukon. For the former an overall accuracy of 74% was achieved, while for the latter average accuracy was 63%. Overall potential was observed for discrimination of some general land cover types. Future work will focused on continued testing of these methods, as part of future investigations under Environment Canada´s Emergency Spatial Pre-SCAT for Arctic Coastal Ecosystems (eSPACE) project.
  • Keywords
    geophysical image processing; image classification; oceanographic techniques; remote sensing by radar; Arctic Coastal Ecosystems; Canadian Arctic shoreline mapping; Environment Canada Emergency Spatial Pre-SCAT; Ivvavik; Northwest Territories; RADARSAT-2 data; Richards Island; Yukon; eSPACE project; general land cover types; object-based approach; object-based classification; optical data; Accuracy; Arctic; Ecosystems; Remote sensing; Sea measurements; Synthetic aperture radar; Videos; Arctic; Classification; Coastal; Object-Based; SAR;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Geoscience and Remote Sensing Symposium (IGARSS), 2014 IEEE International
  • Conference_Location
    Quebec City, QC
  • Type

    conf

  • DOI
    10.1109/IGARSS.2014.6947653
  • Filename
    6947653