Page 1Page 2Page 3Page 4Page 5Page 6Page 7Page 8Page 9GEO/EVS 425/525 Unit 3Composite Images and The ERDAS Imagine Map ComposerThis unit involves two parts, both of which will enable you to present data more clearly than you mighthave thought possible. You will make a composite image based on the Landsat Enhanced ThematicMapper + satellite, and you will learn to use the ERDAS Map Composer. The former includes 7 spectralbands with a resolution of 28.5 meters per pixel and a panchromatic band with a resolution of 14.25meters per pixel. You will merge the two resolutions of the TM imagery and then superimpose somethematic imagery over your merged image. You will then use Map Composer to print your image. Onceyou have completed that, you will make hard copies of images you created in Unit 2. Map Composer is the main mechanism by which images in Imagine are sent to the printer. Its userequires that images be readable within Imagine and that the computer in use be attached to a printer,either locally or on a network. In this class, you will use our networked printer, the Magicolor 6100.You printed an image directly from the viewer in Unit 2, and you may question the need for the MapComposer. There is a great difference between the two modules. When you print from the viewer, yousimply print the image and a logo. You have very little choice. When you use the Map Composer, youhave a great deal of choice of the layout of your map as well as the other things you wish to print alongwith it. The images produced are very high-quality maps and presentation graphics. Maps produced with the Map Composer include continuous and thematic raster layers, vector layers,mixed images, etc. These maps can include text, legends, scale bars, grid lines, tick marks, borders,symbols, etc. You can select from over 16 million colors, multiple line styles, and over 60 text fonts. Making your Composite Map: Fusing Imagery of Different ResolutionYou are going to make a rather interesting composite map for your first composition. Your first step will beto fuse thematic mapper imagery of different resolutions; you will then superimpose some simple thematicimages over your fused image. On the Q drive, you can find two TM images of the Shaker Heightsquadrangle taken in October, 1999. The panchromatic image is Shaker_L7_PC; the multispectral imageis Shaker_L7_MS. Open two viewers and click on Session -> Tile Viewers to line them up next to eachother. Open each image in a viewer so that you can compare them. Which has the most resolution? Canyou see things in one that you cannot in the other? Rather, what can you see in the multispectral imagethat you cannot see in the panchromatic image, and vice versa? To fuse the two images, click on Interpreter -> Spatial Enhancement -> Resolution Merge. Your high-resolution file is Shaker_L7_PC; your multispectral file is Shaker_L7_MS. Give your output file a suitablename. You can accept the default values for the details of the merge, but you might want to try out someof them to see what they do. Click on OK. Look at your result. What did you do?Making your Composite Map: Getting Information from Thematic MapsYour next step will be to obtain information from two images showing quantitative environmental dataabout the Shaker Heights quadrangle. The two images are ShakerImpervious, which estimatespercentage of impervious surface on a scale of 0 - 100% and ShakerCanopy, which estimates percentageof tree canopy on a scale of 0 - 100%. Imagine that you have been hired to do an assessment ofenvironmentally significant areas in the Shaker Heights quadrangle, and that your client wants to knowwhich areas are extremely impervious (i.e. > 75% impervious surface), which have very high canopycoverage (i.e. > 75% canopy closure), and which have high canopy coverage (i.e. > 50 but < 75% canopyclosure). You need to convert your continuous quantitative images into thematic qualitative images thatshow the classes of interest. To do this, you click on Interpreter -> GIS Analysis -> Recode. You will needto do this for each of your two input images. Your input images will be ShakerCanopy andShakerImpervious, respectively; your output images can be anything you want them to be. To convertyour images, you click on the Set Up Recode button after you enter the name of the input image. You willFigure 1: Basic setup ofMap Composition note that the row values are the attribute values; they range from 0 to 100%. Using your mouse in the“row” column, highlight 0 through 50 for the canopy coverage image and change the value to ‘0'. Youhighlight a range of values by holding down the left mouse button and dragging the cursor from 0 to 50. Inthe same way, change values 51 through 75 to ‘1' and 76 through 100 to ‘2'. Click on OK to recode theimage and create your thematic map of canopy cover. Finally, use your mouse on the impervious-surfaceimage to change values 0 through 75 to ‘0' and 76 through 100 to ‘3'. Create your thematic map ofimpervious surface. After you have finished with both maps, you have created two thematic images. The first has canopycover zones labeled 1 and 2, for those of interest, and 0 for “other” areas of the quadrangle. The secondhas an impervious surface zone labeled 3, for the zone of interest, and 0 for “other” areas of thequadrangle. There is no overlap between zones of interest for canopy cover and impervious surface. That is, a specific pixel is of interest either because its canopy cover is high or its impervious surface ishigh. No pixel is high in both. Your next step will be to create a single image combining the information in the two images you have justcreated. To do this, click on Interpreter -> Utilities -> Operators. Your two input files are the two them aticimages you have just created. Give your output file a suitable name. Choose ‘+’ for your operator (it isthe default), and choose “unsigned 2-bit” as your output type. Click on OK to create your compositethematic image. Look at it when you are done. What did the “operator” routine actually do? Because youused the ‘+’ operator, it added the values of each pixel in the input images to calculate the value of thecorresponding pixel in the output im age. This is called “Map Algebra.” W e will have more to say about itas the semester progresses. Because there is no overlap in the areas of interest, a pixel with