Posts

Showing posts from November, 2024

Unsupervised & Supervised Classification Lab 5

Image
This week's lab focused on unsupervised vs. supervised image classification. Image classification is the process of giving pixels in an image unique values. Unsupervised classifies without predefined training data, while supervised classifies guided by known training samples.  The map below shows  a supervised classification of the city of Germantown, Maryland. It was created by using the inquire(legacy), the growing properties, and signature editor features in ERDAS Imagine with the given coordinates. The map shows different types of land use and their coverage. The unique classes chosen were Urban, Road, Deciduous Forest, Mixed Forest, Fallow, Agriculture, and Water. The area of each is shown in the legend in acres. A inset map of the distance image map is also included.

Spatial Enhancement, Multispectral Data, and Band Indices Lab 4

Image
 This lab focused on spatial enhancements which is a preprocessing step in preparing imagery for use. Enhancements modify data and imagery to help users interpret the data beyond what is initially apparent by using a variety of methods. This lab also focused on how to view properties of multispectral data and create band indices.  ArcGIS and ERDAS Imagine were used in this lab to examine an image of the surrounding areas of the Olympic Mountains in Washington State. Most tasks required to complete this lab can be done in either program however using both made the lab faster and was able to have more thorough information. Descriptions of three features were given and the task was to identify that feature within the given image by examining the histogram, examining the image as grayscale, examining the image as multispectral, and using the inquire cursor to find brightness value. As well as use a different band combination for each map. The first feature to identify was: In Laye...

ERDAS Imagine and Digital Data Lab 3

Image
This week was an introduction to ERDAS Imagine 2011 and how to navigate it. I found ERDAS Imagine fairly easy to use and navigate. There are some similarities between ERDAS Imagine and ArcGIS Pro. An image of forested land in Washington State was used to create a map. The Inquire Box tool was used to create the subset map found in the map layout below. A column Area was added to the attribute table to obtain the area of each feature in hectares. The image of the map was then opened in ArcGIS to create a map layout found below. The name of each class was changed to show the hectares of each.