![]() ![]() Using the default mosaic type (“LAST”) the raster that is added first will appear at the back of the mosaic and the raster that is added last will appear on top. This extracted image will be used in the final mosaic rather than the original raster.įigure 2: The orange polygon will be used to extract only the covered area from the raster for use in the mosaic.Ĭreating the mosaic itself is as simple as loading your rasters into the Mosaic tool in the correct order. Once the polygon is drawn, the Extract by Mask tool will create a new raster from the areas of the image under the polygon. If both the top and bottom images are displayed, peeling back the top image with the Swipe tool on the Effects toolbar can help you find areas where the seam will be discrete. Rather than having straight edges which cut across features, the edges of the polygon gradually wind their way across the image following places of natural color and texture change such as edges of vegetation and patterns in the rocks on the beach. Here, I created a polygon feature class and drew a polygon (a portion of which is shown in translucent orange in Figure 2) over the portions of the top image I wanted to keep. Here the seam is obvious primarily due to discontinuities from warping in the images which was not resolved by georeferencing, so color balancing isn’t required for these particular images.Ī simple way to conceal a seam is to move it somewhere else in the image where it will be less noticeable. ![]() ![]() Color balancing can also be helpful when disguising seams caused by variation in the color, contrast and brightness of the images. Seams (areas where two overlapping images meet) that are obvious can be visually distracting, as well as disrupt classifications and other raster analyses. By re-arranging their order I can see that, although the areas of overlap are obvious regardless of the order, the overlap will be easier to hide in the north due to the patterns observable on the beach.įigure 1: The seam between the two images is quite obvious due to changes in lighting and image quality as well as discontinuities along the shore and vegetation. In this case, I have only two rasters, one which covers the southern portion of the island, and one which covers the north. Take into consideration the colors, lighting, how the features changed as the images were recorded, and the accuracy of the georeferencing in different regions of the imagery. Opening all of your images in ArcMap and playing with their order can help you decide which images should overlay others. Making a good mosaic requires some planning. For the purposes of this project, I selected the geoprocessing Data Management tool Mosaic tool, but I suggest reading the help topic What mosaicking tools are available in ArcGIS? for guidance in selecting the method that will work best for your project. ArcMap 10.0 offers many options for mosaicking raster images. Generally, several images will need to be mosaicked (“stitched together”)into a single combined image. ![]() When working with aerial imagery, it is rare that the feature(s) of interest will fit neatly inside a single image. In this post I am going to look at tips and tricks you can apply to improve the quality of your own mosaicked imagery. Now those images must be combined into a single raster. The first step in this process, georeferencing the images, was covered in the previous post. The final goal of the project was to produce a detailed classification of the island’s vegetation from a series of digital aerial photographs. These images are from a project I recently completed looking at the structure of a seabird colony off the coast of Nova Scotia, Canada, and are representative of the less-than-ideal imagery many of us have to work with regularly. This is the second in a series of blog posts that will cover some tips and tricks for performing the following operations on a series of aerial images using ArcGIS 10.0: ![]()
0 Comments
Leave a Reply. |