A HYBRID HIGH RESOLUTION IMAGE CLASSIFICATION METHOD FOR MAPPING EELGRASS DISTRIBUTIONS IN YAQUINA BAY ESTUARY, OREGON

摘要

False-color infrared (CIR) aerial photography of the Yaquina Bay Estuary, Oregon was acquired at extreme low tides and digitally orthorectified with a ground pixel resolution of 20 cm to provide data for intertidal vegetation mapping. Submerged, semi-exposed and exposed eelgrass meadows and macroalgae beds were clearly imaged. Normalized Difference Vegetation Index (NDVI) derived algorithms were developed to classify eelgrass, macroalgae, and non-vegetated areas at the pixel level and assessed at an overall accuracy of ～70%. This method proved to be effective in distinguishing intertidal vegetation from non-vegetated areas but less so in separating vegetation by genera or species. The NDVI derived algorithm was adjusted by individual image and applied to all three bands resulting in a three band intertidal vegetation mosaiced image. The resulting image was reclassified into five statistically distinct classes using unsupervised isoclustering and gridded into a raster GIS. A speckle reduction filter was applied to the resulting grid which was re-tiled and vectorized. A photo-interpreter with orthophotography as an overlay differentiated the vector polygons as eelgrass or macroalgae using software selection tools. This method combines vegetation classification in digital image processing, computeraided drawing in raster-to-vector conversion and photo-interpreter guidance to produce maps less spatially generalized than manual methods and more accurately classified than automated digital processing methods. Remotely sensed data derived with this hybrid classification method have promising uses in areas where data precision and accuracy are required such as in estuarine ecological analyses and coastal resource management.