Lab 3: Vegetation Classification

Goals and Background:

In this lab vegetation is classified in the LAS dataset from labs 1 and 2. The height filter and the manual cleanup are used to complete this task.

Methods:

Part 1: Automatic height filtering of vegetation object points

Automatic height filtering, a very basic filter for categorization of points into height classes is used for the first, most automated portion of this lab. A filter using the Z values of the points above the classified ground points and thresholds for different levels of classification (low, medium, and high) is used to classify these points. This is why it is important to have correct ground classification properly performed before running this filter. Table 1 shows the different classes of vegetation.

Table 1

Figure 1 shows these classifications applied in the filter window.

Figure 1

Part 2: Manual cleanup of the vegetation classification

This part of the lab involved using the same basic filter tool to clean up omission and commission errors as was used in Lab 2 Part 2. Descriptions of these tools can be seen in the previous Lab 2 post.

Results:

Figure 2

Figure 2 shows a neighborhood classified by building (red), ground (orange), and vegetation (green). This figure shows the results of classification performed in Labs 1, 2, and 3.Manual cleanup was also necessary to obtain this result.

Sources:

• LAS, tile index, and metadata for Lake County are from Illinois Geospatial Data ClearingHouse. 
• NAIP imagery is from the United States Department of Agriculture Geospatial Data Gateway. 
• Breaklines are from Chicago Metropolitan Agency for Planning. 
• Lab instruction, data, and other resources were all provided by Dr. Cyril Wilson.