You are here:
LARGE AREA LAND COVER MAPPING THROUGH SCENE-BASED CLASSIFICATION COMPOSITING
Citation:
Guindon, B. AND C M. Edmonds. LARGE AREA LAND COVER MAPPING THROUGH SCENE-BASED CLASSIFICATION COMPOSITING. PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING 68(6):589-596, (2002).
Impact/Purpose:
The primary objectives of this research are to:
- Provide information on the variability in water supply that can be expected under varying climatic conditions. Early efforts will be focused on assembling regional databases for at least two counties (Mecklenberg County and York County) within SEQL region that can be used for water supply generation and model development.
- Develop tools that will help improve our ability to evaluate, study, and model linkages between different types of environmental systems: hydrologic, geomorphic, ecological, and climatic.
- Explore the use of annual and seasonal measurements of large lake surface temperatures as a new ecological indicator of the overall thermal content of those lakes, and construct an estimator of seasonal large lake heat budgets.
Description:
Over the past decade, a number of initiatives have been undertaken to create definitive national and global data sets consisting of precision corrected Landsat MSS and TM scenes. One important application of these data is the derivation of large area land cover products spanning multiple satellite scenes. A popular approach to land cover mapping on this scale involves merging constituent scenes into image mosaics prior to image clustering and cluster labelling thereby eliminating redundant geographic coverage arising from the overlapping image swaths of adjacent orbital tracks. In this paper, arguments are presented to support the view that multiple coverage contains important information that can be used to assess and improve classification performance. A methodology is presented for the creation of large area land cover products through the compositing of independently classified scenes. Statistical analyses of classification consistency between scenes in overlapping regions are employed to both identify mislabelled clusters and to provide a measure of classification confidence for each scene at the cluster level. During classification compositing, confidence measures are used to rationalize conflicting classifications in overlap regions and to create a relative confidence layer, sampled at the pixel level, which characterizes the spatial variation in classification quality over the final product. The procedure is illustrated with results from a synoptic mapping project of the Great Lakes watershed that involved the classification and compositing of 46 Landsat MSS scenes.