Module 10 - Supervised Image Classification
Supervised classification differs from unsupervised methods in that it uses "training sites" to process the classification of images from its spectral values. In this week's laboratory assignment, we exercised an array of sequence steps (using ERDAS Imagine) that are necessary to achieve such results.
The lab scenario entailed creating a supervised classification covering data of the City of Germantown in Maryland. To meet the demands of the city governor's "smart, green, and growing initiative", a current land use map was to be produced.
To create this supervised classification, I created spectral signatures and AOI (Area of Interest) features by using the Grow tool and Polygon tool under the Drawing tab, but first I needed to make sure to add a new, black AOI layer to do so. While creating my signatures, I made sure to adjust the Spectral Euclidean Distance and Neighborhood settings when necessary so that the P, I, H, and A columns were all checked off. During this process, I also recognized and eliminated spectral confusion between signatures in bands by using both the Histogram and Mean Plot tools and kept track of the bands that had the greatest separation among signatures; which were Bands 3, 4, and 5 (as shown in the screenshot below).
To run the classification process using Bands 3, 4, and 5, I did the following: on the Signature Editor window > Edit > Colors > Approximate True Colors > Set Signature Colors. The final steps included: applying a Parametric Rule of "Maximum Likelihood" to create both an Output File and an Output Distance File; merging classes to new, unique values using the Recode tool under Raster tab > Thematic, as well as adding class name and acreage columns to the attribute table. After all changes were saved, I imported my final working image into ArcMap to finalize my cartographic design product.
The lab scenario entailed creating a supervised classification covering data of the City of Germantown in Maryland. To meet the demands of the city governor's "smart, green, and growing initiative", a current land use map was to be produced.
To create this supervised classification, I created spectral signatures and AOI (Area of Interest) features by using the Grow tool and Polygon tool under the Drawing tab, but first I needed to make sure to add a new, black AOI layer to do so. While creating my signatures, I made sure to adjust the Spectral Euclidean Distance and Neighborhood settings when necessary so that the P, I, H, and A columns were all checked off. During this process, I also recognized and eliminated spectral confusion between signatures in bands by using both the Histogram and Mean Plot tools and kept track of the bands that had the greatest separation among signatures; which were Bands 3, 4, and 5 (as shown in the screenshot below).
To run the classification process using Bands 3, 4, and 5, I did the following: on the Signature Editor window > Edit > Colors > Approximate True Colors > Set Signature Colors. The final steps included: applying a Parametric Rule of "Maximum Likelihood" to create both an Output File and an Output Distance File; merging classes to new, unique values using the Recode tool under Raster tab > Thematic, as well as adding class name and acreage columns to the attribute table. After all changes were saved, I imported my final working image into ArcMap to finalize my cartographic design product.
Comments
Post a Comment