Using the Spectral Matrix 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 The image service definition (.ISDef) has been replaced by an improved geodatabase data model—the mosaic dataset—which can be published as an image service using the ArcGIS Server Image extension.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 matrix to a multiband image to affect the spectral values of the output. This can be used, for example, to convert a false color image to a pseudo color image.

Output Band_N = Weight_P * Band_Q

• Band_N is an output band, where N is a number from 1 to the number of output bands.
• Weight_P is a comma-delimited list of weights, one for each input band.
• Band_Q is the input image, where Q is a number from 1 to the number of bands of the input image.

A gain and offset can be applied to each of the output bands, which is often required to ensure the output remains within the allowable range.

This process can be applied to raster datasets with up to four bands.

False color versus pseudo color

A false color image is created by combining three available bands into a red, green, and blue (RGB) display in which the output spectral values each represent red, green, or blue. For example, you could convert a SPOT image that lacks a blue band into what will appear as a true color image by generating a blue band from the other available bands and using the red and green bands that are part of the SPOT image dataset.

A pseudo color image is created from a single-band image (grayscale) or multiband image by applying a transformation to create colors where there are none. For example, a thermal image may be displayed as a pseudo color image using a continuous, three-tone color ramp to produce an image displaying three classifications of heat. A pseudo color image can be created from multispectral images by applying the Spectral Matrix process to determine color bands from other bands.

The following matrix can be used to convert a false color (near infrared, red, green) image to a pseudo color (RGB) image:

0.141  0.949  0.000

0.422  0.000  0.756

0.141  0.105  0.864

For example, on the Band Weight tab, you would type the matrix values as a string for each band, with each value separated by a comma.