Sara's GIS Journal

  This week I completed a supervised land-use classification for Germantown, Maryland. Using ERDAS Imagine, I collected spectral signatures, evaluated their accuracy and recoded the results into eight major land-cover types. The final map created shows the distribution of these land classes, as well as a distance file to highlight areas of potential misclassification. 

  This map represents using multispectral analysis to identify features in an image. This lab taught me how to interpret image histograms and identify features through various digital data. I was able to successfully use ERDAS Imagine to process and visualize satellite imagery, while applying techniques of remote sensing. The final map was created in ArcGIS to display the three features found. 

  This week I learned how to use ERDAS Imagine to view and manipulate imagery and prepare data for mapping in ArcGIS. I learned how to interpret differences in spatial resolution and spectral information. This process taught me how to reveal land cover patterns to prepare accurate maps for analysis, such as the one above showing land classifications for forested lands of Washington State.