About the Classify Pixel process

Legacy:

ArcGIS 10 is the last release of the stand-alone ArcGIS Image Server product. The image service definition (.ISDef) has been replaced by an improved geodatabase data model—the mosaic dataset—which can be published as an image service using the ArcGIS Server Image extension.

This process applies a simple classification of pixels to the raster data used in the image service. The classification of pixels is generally used for feature discrimination. If you want to clearly demarcate pixels based on the darker areas (such as roads or water bodies) and lighter areas (such as buildings or open land) in the displayed image service, you need to apply the Classify Pixel process chain. You should generally apply the Classify Pixel process in conjunction with another process that involves computations, such as NDVI and Image Algebra.

The output of a Classify Pixel process is a single-band, black-and-white image, with the selected pixels that have been included being white (255) and those that have been excluded being black (0).

The ArcGIS Image Server performs the pixel classification in two steps. Each pixel is classified based on whether it passes the Include method first and the Exclude method second. If it passes both tests, its value is defined as 255; otherwise, its value is 0.

Include methods

• All—This includes all the pixels in the raster. No ranges need to be specified, which effectively eliminates this test.
• And—This uses the ranges specified in the Include Range fields to determine the pixels that will be included. A pixel is included only if the pixel's value falls within all the ranges specified.
• Or—This uses the ranges specified in the Include Range fields to determine the pixels that will be included. A pixel is included if the pixel's value falls within any of the ranges specified.

Exclude methods

• None—This specifies that no pixels will be excluded from the raster. No ranges need to be specified, which effectively eliminates this test.
• And—This uses the ranges specified in the Exclude Range fields to determine the pixels that will be excluded. A pixel is excluded only if the pixel's value falls within all the ranges specified.
• Or—This uses the ranges specified in the Exclude Range fields to determine the pixels that will be excluded. A pixel is excluded if the pixel's value falls within any of the ranges specified.

These two methods can be used in combination. This enables a classification in which a pixel is set to white if it has a specified include range but also falls outside a specified exclude range. Note that all ranges specified are inclusive. For example, a range of 4–6 and 10–20 could be defined as include 4–20 and exclude 7–9.

Below, the image on the left is an example of a three-band color image. The image on the right (shown in black and white) has the Classify Pixel process applied to differentiate dark areas of the color image from the lighter areas. In this case, the light areas in the original image are black, and the dark areas, such as roads and water, are white in the output image.

The Classify Pixel process can be used to demarcate the pixels as shown above, using either the Include method or the Exclude method.