Help Sheet for ERDAS 8.3.1 1.

How to compare the spectral properties

• one path: RASTER > Profile tools
• a. spectral profiles: place one or more points on the image to examine the DN values for the pixel in each band.
• b. spatial profiles: draw a line through different surface materials and examine the change in DN value for one band at a time through this space.
• c. surface profile: draw a box over the area of interest and see a 3-D view of the DN values for this space one band at a time.

· Linking Viewers: To enhance your interpretation of the image, ERDAS allows you to geographically link two images. First step - open a new viewer. Load two images. Right click in either viewer and geolink/unlink. Follow the instructions on the screen. Zoom in and out and note the relationship between the two images. From either viewer click on the inquire curser (the big + sign icon). This will allow you to examine the same pixel in both viewers. It will also state the DN values of the linked pixels.

Examining spectral space: One method in ERDAS of exploring spectral properties is by creating feature space layers. This function can be found in classifier > signature editor under feature. This allows the user to create a scatterplot where the data file values of one band have been plotted against the data file values of another band. This can only be done in a two dimensional histogram, but theoretically our data has 6 dimensions of spectral space. In the window titled create feature space images you must input the image you want to evaluate and click output to viewer. You may choose from a list of band pairs at the bottom of the window under feature space layers. From the signature editor dialogue choose feature > view >select viewer. Click inside the feature space viewer. Then select feature > view > linked cursors and click link. Then click inside the viewer with your image. The viewers are now linked.

How to subset and mosaic images:

¨ There are multiple ways to clip (SUBSET) an image, many of them use the subset command. The method that I have used and found to work the fastest is the following:

• Start with the image you want to clip in a viewer
• Click on UTILITY and select INQUIRE BOX
• Move the box and enlarge it include the area you want to clip
• Click the DATAPREP button, and select SUBSET IMAGE
• Fill out the options selecting the image input and output, and click on FROM INQUIRE BOX button. This will make a .img file that contains the area, layers, and data type you selected.

Here is one method to (MOSAIC) multiple Images (individual blocks) together. Once you have all your clipped images, you will want to put them back together for your unsupervised classification. Do the following:

• Click the DATA PREP button
• Select MOSAIC IMAGE
• Click EDIT and ADD IMAGE or use the button
• Add all your clipped images
• Then click PROCESS and RUN MOSAIC
• This will make a new .img file with all your clipped images included.

How to create signature files

A signature is a set of statistics that is created when clustering and defines a training sample or cluster. The signature is used in the classification process. Each signature corresponds to a GIS class that is created from the signatures with a classification decision rule.

Using training sites (clustering in a supervised manner): Training sites can be identified as Areas Of Interest.

• Under AOI > tools, use one of the many buttons to choose a region (square, polygon, single pixel etc.).
• Then under classifier > signature editor use the create new signature from AOI button to add this region's signature. This signature defines a potential class. Note that it is also possible to replace current signatures and merge signatures in the editor as well.
• A set of signatures can be saved in one signature file (.sig) using file > save as. This file can then be used to do a supervised classification.

Unsupervised Classification: a signature file is automatically created when an unsupervised classification is completed.

• You must name the .sig file before the image is classified, and it will contain as many signatures as there are classes.
• You decide the # of classes.
• This .sig file can then be opened in the signature editor.
• The signatures can be merged with other signature files, as we will do with the supervised training site signatures, making this a hybrid approach.

How To Evaluate Classification

• file >open > raster layer and display bands 4,5,3 respectively
• from the viewer tool bar open a classified raster layer, make sure the "clear display" option is turned off from the "raster options tab" found from that dialog box
• select Raster > Attributes from the viewer, the editor will be displayed
• NOTE: you can edit the way this editor appears by selecting Edit > Column Properties
• Start by setting opacity for all classes to "0". In the Raster Attribute Editor, click on the word OPACITY at the top of the Opacity column, then right-click-hold on the word Opacity and select formula from the column option menu, click on the "0" (the zero on the number pad); and click on Apply.
• Now change the color for class 1 to something like Yellow or red so it’s easier to see, … click on the color patch and change the color.
• Change the opacity for class 1 in the Cell Array to 1 and press return, this class will be shown in the viewer.
• From the viewer menu bar select Utility |> Flicker … the dialog opens … turn on the auto mode.
• .The flashing pixels are the pixels in this class, click on the Class_Name in the editor and give it meaningful name and assign it a meaningful color
• Repeat for the other classes.