Laboratory Exercise 7

Aerial Imagery Scanning and Geometric Correction
(or jump to Digital Elevation Model)

Step 1. Become Familiar with the Principles and Practices of Image Scanning

Step 2. Scan Aerial Photograph and Generate a Tagged Interchange Format File

Step 3. Become Familiar with Principles of Geometric Correction and Resampling.

Step 4. Acquire coordinate information for Ground Control Points (GCPs) to be used in

geometric correction.

You have access to an aerial photograph already-scanned of the greater University of Hartford campus. This image is, of course, an unrectified aerial photograph. Follow the instructions from you instructor or lab assistant in placing this Idrisi formatted image, both AERIAL.IMG and AERIAL.DOC on your hard drive in your working Idrisi directory. You have access, also, to an Idrisi-formatted vector coverage for roads, named ROADS.VEC and its accompanying documentation file, ROADS.DVC. Similarly, place these two files in your working directory.

The ROADS.VEC was obtained from a State Plane formatted ArcInfo coverage located on MAGIC at The University of Connecticut. The data were:

The final roads vector layer is in UTM, Zone 18, NAD 27, and in units of meters.

You will use a DISPLAY of this roads vector layer, along with a DISPLAY of the unrectified raster image, to select pairwise Ground Control Points (GCPs) from both. The roads layer will provide the real world coordinates by which the aerial photograph will be geometrically corrected and geographically projected.

Record the GCP descriptions (to serve as personal annotation and referencing) and real world UTM easting and northing (x- and y-) coordinates for ground control points (GCPs) derived from the roads layer. You will have to display this vector coverage and use the Idrisi inquire cursor to get the UTM coordinates. A set of 8 to 10 quality points is desirable. Leave room in this table for the scanned image coordinates to be added later.

Following is an example of the partial table for recording these GCPs (using some locations for the greater UConn area as examples):

NW corner 195-N.Eagleville Rd. 728401 4632299
SW corner N.Eagleville & Hillside727694 4632063
NE Corner Hillside & Gilbert728286 4631407

Step 5. Use Idrisi for Windows to Perform Geometric Correction

Note: This first step of importing the scanned aerial photo from a TIFF into an Idrisi image has already been performed for you, but the steps are shown below for your information.

Step 6. Visually Assess the Quality of the Corrected Aerial Photograph

Digital Elevation Model


Later this semester, we will be conducting a multicriteria evaluation, multiobjective land allocation (MCE/MOLA) exercise which will address the selection of suitable sites for the allocation of a specified number of hectares for residential, commercial, and industrial development, as well as open space and passive recreation. As we will see, MOLA strives to resolve conflicts among competing land uses and siting perspectives (e.g., real estate developer versus engineer versus conservationist, etc.). MCE enables the combination of a number of constraints and factors, such as those based on existing land use, potential impacts to water and wetlands, proximity to or distance away from certain type of land uses and covers (i.e., buffering), location of rare or endangered species, topographic slope and aspect, and possible many others. Whereas you will be provided with a database for the study, it would be desirable for you to see and experience what takes place in constructing at least a few of the data layers with which you will work later on. The first of these is elevation.

In Idrisi for Windows Tutorial Exercise 8 you were introduced to digital elevation model (DEM) data and information that can be extracted from them, such as degree of slope, aspect orientation, and hill shading, using the SURFACE program. In this activity, you will learn a bit more about digital elevation models, download DEM data for the greater Mansfield area, examine some of the quality problems associated with these data, apply an elementary data cleaning procedure, and extract topographic-related information using Idrisi 's SURFACE program.


DEM Models

7.5 Minute Quad 30-meter DEMs

1 Degree Map Sheet DEMs

Put on Your Thinking Cap

In both the aspect and analytical hill shading layers, the noise in the original DEM data is amplified. Why do you think this is so much more apparent than in the original DEMs themselves?

The Future

You can continue to examine and experiment with this DEM, maybe using the ORTHO program to perform some 3D renditions, but later this semester, when we begin the MCE/MOLA project, you will work with a more highly sanitized version of these data.