By classifying the image based on more specific land-use types, you will create a more accurate classification. If you attempted to classify the segmented image into only pervious and impervious surfaces, the classification would be too generalized and likely have many errors. Then, you will add subclasses to each class that represent types of land cover. The training samples inform the classification tool about the variety of spectral characteristics that each land cover can exhibit.įirst, you'll modify the default schema to contain two parent classes: Impervious and Pervious. Training samples are polygons that represent distinct sample areas of the different land-cover types in the imagery. To perform a supervised classification, you need training samples.
Then, you will reclassify those land-use types into either pervious or impervious surfaces.
#Undo on pcswmm software#
(An unsupervised classification, by contrast, relies on the software to decide classifications based on algorithms.) You'll first classify the image into broad land-use types, such as vegetation or roads. A supervised classification is based on user-defined training samples, which indicate what types of pixels or segments should be classified in what way. In this section, you'll set up the classification of the image.
Because you want to distinguish between pervious and impervious surfaces (which generally have very different spectral signatures), you will use a lower value. A higher value means that pixels must be more similar to be grouped together, creating a higher number of segments. It sets the level of importance given to spectral differences between pixels on a scale of 1 to 20. For instance, the roof of each house is generalized as a single segment. Compare the nonsegmented image (left) with the segmented image (right). The optimal number of segments and the range of pixels grouped into a segment change depending on the image size and the intended use of the image.īelow is an example of what segmentation looks like. Instead of classifying thousands of pixels with unique spectral signatures, you will classify a much smaller number of segments. Doing so will generalize the image and make it easier to classify. The Segmentation process groups adjacent pixels with similar spectral characteristics into segments. You'll now choose the parameters for Segmentation. The next page of the Image Classification Wizard focuses on segmentation. This data includes imagery of the study area and land parcel features.
#Undo on pcswmm download#
To get started, you'll download data supplied by the local government of Louisville, Kentucky. Then, you will reclassify those land-use types intoīefore you classify the imagery, you will change the band combination to distinguish features clearly. You will firstĬlassify the image into broad land-use types, such as roofs or Once you segment the imagery, you will performĪ supervised classification of the segments. Which will generalize the image and significantly reduce the number Instead, you'll group pixels into segments, Which almost every pixel has a unique combination of spectralĬharacteristics, you are likely to encounter errors and However, if you try to classify an image in Using multispectral imagery for this kind of classification works well because each land-use type tends to have unique spectral characteristics, also called spectral signature. Pervious surfaces include vegetation, water bodies, and bare soil. Impervious surfaces are generally human-made: buildings, roads, parking lots, brick, or asphalt. To determine which parts of the ground are pervious and impervious, you will classify the imagery into land-use types.
0 kommentar(er)
