Supervised Image Classification

The top image has been reclassified (supervised)by manually selecting pixels that represent certain features. A few selections were made per feature type (water, residential, agricultural etc...) Common feature classes were then recoded to combine them into more general groups. These groups were then color coded to match their feature type (i.e. water=blue etc....).

I thought the procedure went smooth, however, the results were rather confusing. There seems to be a problem with my classifying of grass and residential areas. The problem may be the Region Growing Properties, specifically the Spectral Euclidean Distance. I am still a bit confused about this function.

Unsupervised Image Classification

The top image is the reclassified (unsupervised) image. The bottom image is the original. The image was reclassified into 15 classes. These classes were then recoded into similar and more general class types (i.e. all water, agriculture etc......). Appropriate colors were then assigned to each land use type.

I encountered issues post reclassification with some land class distinctions....mainly cleared land, and crop land. These issues could have been addressed, but were beyond the scope of this lab.

Image to Map Rectification

By linking an Orthophoto to its corresponding Map image (Image to Map Rectification), cultural features can now be used in a thematic context (related to other data in a GIS). These cultural features now have actual map coordinates, rather than pixel x, y locations, resulting in accurate distance and bearing calculations.
Errors can occur with the GCP method of rectification. An image may lack any discernible features to use as control points. User error in the placement of GCP's (misinterpretation of features). The Orthophoto is distorted beyond the point of rectifying.

DHS Infrastructure Protection Lab

Thermal Infrared Remote Sensing

Provide a brief description of why the following features appear as they do in this thermal infrared image:

Roads: Roads appear lighter due to their radiant temperatures and heat capacity. Road material absorbs more heat and remains warm through the night.

Natural and man-made vegetation: Vegetation appearance will differ depending on cover ( i.e. grass vs. sand) and moisture content of the soil (i.e. watering differences among neighbors and watering amounts in the front yard vs. backyard). The shade differences in the image could be a combination of both.

Sidewalks and Patios: The heating effects of patios and sidewalks should be similar to roads. Variations will be caused by surface types and materials.

Storage sheds in backyards: Sheds appear dark because of their low radiant temperatures. They lack an active or reflective heat source.

Automobiles: Cars will resemble sheds, except if the engine is running (or had been running prior to the image). Unlike sheds, cars do have a a heat source. The heat emitted from the engine will appear lighter.

Bright spots on many of the roofs: Vents on the rooftops will appear lighter than the surrounding roof, due to their higher radiant temperatures. This could be caused by their reflective material, but more likely, their heat source.

Houston Crime Analysis

Module 2 Lab -Spot Panchromatic & SPOT Multispectral

Compare the SPOT panchromatic image and the multispectral mode display. Note any difference?

The two obvious differences are the color and resolution. The Pan image is a single band image (band 1) and the Multispectral image, like the name suggests, incorporates three spectral bands (bands 4,3,2). The Panchromatic image is at 10 meter ground resolution and the Multispectral image has a resolution of 20 meters. The displays also differ in their pixels per line values. Spot Panchromatic has 6,000 pixels per line and the Multispectral has 3000. The Panchromatic displays higher resolution and pixels per line producing a sharper image.