VOLUME 73, NUMBER 8
PHOTOGRAMMETRIC ENGINEERING & REMOTE
SENSING
JOURNAL OF THE AMERICAN SOCIETY FOR PHOTOGRAMMETRY
AND REMOTE SENSING
This series of images of Redfish Bay, Texas provided by Fugro EarthData, Inc.
demonstrates the image classification steps from digital orthoimage to GIS map. This benthic (seafloor) mapping project was performed by Fugro EarthData for the NOAA Coastal Services Center and will support the Texas Seagrass Monitoring Program. The bottom image, a true color digital ortho, displays submerged aquatic vegetation (seagrass and macroalgae), oyster reefs, mangroves and upland islands. The aerial imagery was collected at a 1m pixel resolution in November 2004 using a Leica ADS40 camera as part of the USDA National Agriculture Imaging Program. The middle image depicts habitat polygons, created from the ortho using Definiens Professional, overlaid on top of the ortho. The top image showcases the final benthic habitat map, which incorporates field verification data. This map will be used to help locate, monitor, and protect seagrass beds along the southern Texas coast as well as support other coastal management and environmental issues. For more information call +1-301-948-8550 or email info@earthdata.com.
The relationship between classification accuracy and within class variances is investigated showing that within class variances
are a function of image resolution.
An examination of several methods to spatially aggregate biophysical
measurements collected in field surveys to the scale of moderate spatial resolution remotely sensed data.
Land -use and land-cover classification in an urban rural fringe of the San Francisco Bay Area using EO-1 Hyperion imagery is compared with that using EO-1 ALI imagery, and the application of a computationally efficient segmentation-based feature reduction approach.