Photo Interpretation and Remote Sensing: Lab 5

Supervised Classification: Germantown, MD

 

Final Layout


It took me a little toggling around to figure out the best band combination, but after playing with different ones that highlighted distinct vegetation types, I settled on R-4, G-5, B-6. This band combo clearly contrasts agricultural, grass, and forested lands, making it ideal for the image in question. For the supervised classification I opted to create signatures from “seed” using the inquire cursor in ERDAS imagine.


Area calculated in ERDAS


Comments

Post a Comment

Popular posts from this blog

Module 5: Exploring and Manipulating Data

Image
 Module 5: Exploring and Manipulating Data  This module focused on the implementation of cursors and dictionaries within newly created geodatabases. We learned a lot of different functions and techniques applying to geospatial data. The most notable problem I encountered this week was compiling the search cursor queries into one script. One of the steps in the project was to create a Search Cursor for the cities layer to find the Name, Population (POP_2000), and Feature of all County Seat Cities. Initially, I could only get the cursor to run by dividing each of these categories into a separate cursor file, generating a long list of either city names, populations, and features, respectively. This was for all intents and purposes functional, but was practically illegible and a definite eyesore. By trial and error I was able to create separate variables, and vitally separate print row.getValue statements for each of my fields to make the same script generate all inquiries simulta...
Read more