Object Based Classification Using Arc Pro

In a previous post we engaged in object based classification of multi-band aerial imagery. In this post we are going to engage in object based classification of an RGB image. Although this image is not of an airport, the classification we will engage in is very much something that those managing and maintaining an airport are interested in. Maintenance of airport runways is very much like that of maintaining a freeway or a highway, but with objects that weigh much more, and put more stress on certain areas. Therefore, finding and addressing issues with cracks is very important, along with understanding how much area is permeable vs. impermeable from a drainage standpoint.

In todays post we are going to do a classification analysis on an RGB image where we first create a series of landcover categories. We will then simplify those categories into permeable and impermeable. We will then perform a field calculation to determine how many square meters each of those landcover maintains. Following that, we are going to experiment with classifying and quantifying pavement cracks on a subset of the image. Here we will want to end up with 3 classes, with one being vegetation, another pavement, and another as cracks. We will then examine how accurate the classification is. With regard to the field calculations, I will be referencing another blog post by a previous student that can be found here. 

With getting the data ready, I copied over the file that was provided to us by our professor. The data that was collected and that we will be classifying is over the Tippecanoe County Amphitheater. We have engaged in the same operation with a different dataset in previous blog post that you can view here. The steps will be the same, just with a different set of data.

• The first thing we want to do is classify the country park mosaic. We will engage in object classification where we will create classes which include Asphalt, Concrete, Buildings, Forest Deciduous, Grass, and Shadows. These will each be under a parent class of either Impervious or Pervious. In order to do this:
• Right click NLCD2011 and select Add New Class
• Name: Impervious
• Value: 20
• Color: Gray 30%
• Click Okay
• Right click NLCD2011 and select Add New Class
• Name: Pervious 
• Value: 40
• Color: Quetzal Green
• Click Okay

Once those are created, create each subclass under those parent classes. Once those are created, collect samples of each class using the polygon button. Once you have all your samples collected, collapse them using the collapse button. Now that we have created training samples, we will select the classification method. Each method uses different statistical process involving the training samples. Once the classification has finished, we need to reference the post from a previous student that I mentioned before to calculate the area. 

• You want to begin by using the "Reclassify" tool 
• The class values are kept the same, the purpose of this is to acquire the pixel count

Figure 128: Shows Reclassify Tool For Pixel Count

    The output was created to have the same classification as before, only this time, an attribute table containing pixel counts for each class was generated. Within each attribute table a new column was created and named "Area". The column was added to the attribute table with <NULL>  values in its rows. To generate the area for each class, the ground sampling distance (GSD) was obtained and then, using the "Calculate Field" tool in the attribute table window, the area was calculated by multiplying the cell size GSD, the area for each class was calculated in square meters. Once the classification and area calculation has finished, I created a cartographic map to show the Tippecanoe County Amphitheater.

Figure 129: Classification Of Tippecanoe Country Amphitheater 

The next page is the Merge Classes page, we will use this page to merge the subclasses into their parent classes. While the classes are important for an accurate classification, we are interested in whether it’s pervious or impervious. 

• For each new class, in the New Class column, select either Pervious or Impervious
• Click Run

Figure 130: Tippecanoe Country Amphitheater Impervious and Pervious

After processing the entire Amphitheater, we are now going to focus on a specific part of the driveway and classify the cracks in the road. The steps are the exact same as we did before, except the classes we are going to create are: Pavement, Cracks, and Vegetation. After finishing the classification and the area calculation, I created the map below:

Figure 131: Tippecanoe Country Amphitheater Road Crack