Georeferencing is a technique of spatial positioning by means of which a digital image is assigned a reference system based on known coordinates.
Some raster information is not associated with a reference system, such as maps, scanned topographic maps, aerial images, etc. To link an image to a known coordinate system, you must perform the georeferencing process, which is explained below.
In the georeferencing of an aerial image, once the ArcMap application is executed, click on the Add Data icon in the Standard bar load the image (Figure 5).
The directory is scanned and the raster file that is to be georeferenced is selected (for the present example it is possible to capture and / or save the image of Figure 4 in tif, jpg, png format), ignoring the unknown spatial reference message. In topographic maps, several control points can generally be identified through the coordinates shown on the X and Y axes. In this example, two known points P1 (UTMX: 691194; UTMY: 9532767) and P2 (UTMX: 691895; UTMY: 9532347) are used.
Activate the Georeferencing toolbar (Figure 6): If not found use the Customize > Toolbars menu.
Make sure that the added image (georeference_image.jpg) is inside the bar. If you cannot see the image in the current screen view, click Georeferencing > Fit To Display.
Now click on the Add Control Points icon; this tool allows you to adjust the image spatially by means of known points. The first step is to mark a known point on the image (P1). Once the cursor is positioned, it turns green, then right-click and select Input X and Y to enter the XY coordinates (Figure 7). The process is repeated for the second point (P2) and for the other points that need to be added.
To finish with georeferencing after acceptable adjustment of the control points, go to Georeferencing > Update Georeferencing. This will create auxiliary files to the current image containing the transformations made. If Rectify is selected, the changes are saved and a new georreferenced image is created.
If you do not have the coordinates of the control points on the ground, but you do have reference elements such as roads, buildings, trees, etc., it is still possible that these elements can be used for georeferencing. For example, the georeferencing of an image based on a track shapefile is done by identifying a common point (intersection) in both layers (call up Zoom To Layer with right click on the track layer, then call up the Fit To Display image from the Georreferencing bar). To create a link, first click on the image to mark the starting point (green cross in Figure 8a), a second click is marked at the actual point on the terrain (grey cross in Figure 8a). For the following points, the process is repeated in a similar way. The blue line joining the checkpoints means that an associated link has been created between a point in the image and the real world (Figure 8b). The adjustment of the two layers is done automatically after each point association.
The number of control points is not strictly reflected in the quality of the result, and the important thing is to make an adequate distribution. Try not to concentrate the control points in a single sector (Figure 9b) but distribute them homogeneously within the image (Figure 9a). The more points you can apply, the more complicated the transformations (polynomial, spline, adjustment or projective) you can use to determine the proper location of the coordinates for each cell of the raster image.
You want to learn an ArcGIS 10 Course following this manual + online + certificate = 15.99 USD