An Evaluation of Image Processing Techniques for Mapping Seagrass from IKONOS Satellite Imagery in Springs Coast Florida and Evaluating the Effectiveness of These Techniques in Replacing Traditional Photo-Interpretation Methods

Rene Dieter Baumstark

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, Department of Environmental Science, Policy and Geography, College of Arts and Science. University of South Florida St. Petersburg

Abstract

Seagrass mapping in Florida provides important information about the spatial distribution and density of seagrass habitat in coastal waters. New image processing and satellite technologies present the opportunity to leverage quantitative, objective and cost effective techniques that have the potential to improve upon traditional aerial photography photo-interpretation techniques. Pan-sharpening, light attenuation correction and spatial/spectral image processing techniques were evaluated for their effectiveness in mapping seagrass from an IKONOS satellite image of Springs Coast, Florida. Pan-sharpening techniques, which are limited by the quality of the panchromatic band, were found to alter original multi-spectral pixel values in a statistically significant way and did not improve the photo-interpretability of this scene. The application of depth variant light attenuation corrections improved spectral classification results but was limited by the quality of bathymetric data. A combination of pixel classification and image segmentation techniques provided a seagrass density index map that represented seagrass density and distribution with high fidelity and overall accuracy (77%) comparable to photo-interpretation techniques. Satellite imagery based image processing techniques were found to provide a more comparable and cost effective data source and mapping technique than aerial photography for seagrass mapping in Springs Coast.