Pan Sharpen Imagery

Pan Sharpening is an analysis method by which lower resolution color imagery can be sharpened (increased in resolution) using a higher resolution panchromatic image. This tool can be found in the Analysis menu.

Input Color image and Panchromatic image, produce a Pan-Sharpened Image (IHS) .

Pan Sharpening is typically done with remotely sensed imagery, where a high resolution panchromatic image was collected simultaneous to lower resolution color imagery. For example Landsat8 and Spot5 both collect high resolution Panchromatic bands. The required inputs to perform pan sharpening are a single band high resolution image and a lower resolution 3 or 4 band image (typically RGB or RGBI).


Specify the algorithm to be used for Pan Sharpening. Each method takes every pixel from the panchromatic image (or as specified in the sample spacing) and determines a value for each band using the color image. The band values are normalized based on the image histogram (min and max values) and any contrast adjustment settings that have been applied. The algorithm used will be stored in the metadata of the pan-sharpened image.

The resampling method for determining the values used in the algorithm at each output pixel location come from the layer options. For imagery the default setting is Nearest Neighbor.

Color Layer

Specify the color image to pan-sharpen. This can be a 3 or 4 band multi-band image, or an RGB image.

Pan Layer

Specify the panchromatic layer. This must be a single band image. Typically pan sharpening is performed with a panchromatic image collected simultaneous to a color image.

Panchromatic imagery is collected by sensing all wavelengths of visible light, as opposed to color imagery where bands are collected separately based on the wavelength.

Band Weights

Band weights are factored into the chosen algorithm. Each method uses a sum or average calculation of the RGB (I) color band values when calculating the output bands. Different weighting can be applied by choosing to use custom band weights. The band weight values should add up to 1, however compensation will be done if they do not.

Band weighting can be applied to the specified pan sharpening algorithm to correct color, or enhance certain parts of the image.