Module 4 - Image Preprocessing: Spatial & Spectral Enhancements and Band Indices - Spatial Enhancement, Multispectral Data, and Band Indices Lab

 This module's lab exercise continued my practice with ERDAS Imagine and how aerial photos can be manipulated and presented using the program. The lab module covered a range of topics including data acquirement, image formatting, essential image filtering, and what tools are necessary for feature identification and proper data presentation.

One of the main focuses of this lab included how to filter aerial photos using ERDAS Imagine to best view certain image patterns. This is also known as spatial enhancement. I learned how kernels are used to create low pass, high pass, and sharpen filters to change an image. I learned about other common filters, however the three aforementioned were actually conducted during this lab. 

Following the topic of filters, I learned how to use and interpret image histograms for aerial photos. This included a robust look into pixel values and what spikes in the histogram mean, as well as how to adjust an image using breakpoints on the histogram. I practiced this in both ERDAS Imagine and ArcGIS Pro. 

Next, I practiced changing how an image is displayed, both in multispectral and panchromatic views. I learned key band ratios common for representing data in specific ways, such as True Color, TM False Color IR, and TM False Natural Color. I also looked at key image metadata in these views using the Inquire tool to view features like LUT values per RGB band. 

The lab assignment culminated into a final exercise that tied in all of these image processing tools and concepts aforementioned. For this exercise, I was given a base image and tasked to identify three features based on specific qualities detailed in the lab exercise. To complete this assignment I would to need to examine histograms of the image to determine shapes and patterns in the data, examine the image in various layers using grayscale to determine pixel brightness patterns, use the Inquire Cursor to find exact brightness values of particular areas, and finally visually examine and present the determined feature using a multispectral band combination that promoted contrast. The images below are the three features I identified and created a map for using a subset image from ERDAS Imagine and Layout from ArcGIS Pro. An image caption is included in each layout.