TIDAL WETLAND CLASSIFICATION FROM LANDSAT IMAGERY USING AN INTEGRATED PIXEL-BASED AND OBJECT-BASED CLASSIFICATION APPROACH

摘要

The tidal wetlands within the Long Island Sound estuary serve a critical role in maintaining the health of the Sound. Over the past two centuries, there has been significant disturbance and loss of tidal wetlands along the Sound due primarily to anthropogenic activities. Researchers at the University of Connecticut and Wesleyan University are continuing on the second year of a two year project to document the extent and vegetative composition of coastal marshes using moderate resolution Landsat ETM+ and Terra ASTER satellite imagery and high resolution QuickBird satellite and Leica ADS40 aerial imagery in conjunction with in situ field measurements of plant spectra. This paper will detail research to classify tidal wetlands throughout Long Island Sound from Landsat satellite imagery. The goal of this portion of the project was to produce an accurate base map that identifies the location of tidal wetlands. An integrated classification approach which uses both pixel-based and object-based classification techniques was utilized. The classification serves as a base map to compare with subsequent dates of imagery to monitor any changes in tidal wetland extent and also compared with existing land cover maps to identify any upland changes in close proximity to the wetlands that could cause potential detrimental impacts to the tidal marsh system. The results of this research will provide a beneficial tool for coastal wetland management and monitoring along the Long Island Sound estuary.