USGS MIPS Rgb2his Documentation

USGS MIPS Rgb2his Documentation

This program converts three images from an RGB color composite to hue, intensity, and saturation images. The algorithm was obtained from Fundamentals of Interactive Computer Graphics by J.D. Foley and A. Van Dam, pages 618-619.

The hue image represents the color of the RGB values. It will contain 8-bit values from 0 to 255. The intensity represents the brightness of the RGB values where 0 is black and 255 is white.

The saturation image will contain floating point values from 0 to 1 where the value represents the percentage of saturation. Any RGB value with a saturation of 0 represents a greytone. As the intensity increases or decreases the greytone shifts toward white and black, respectively. A saturation of 1 is used to represent pure or more intense colors while mid-range saturation values represent softer colors such as pastels.

Files

This program requires as input three unsigned 8-bit image files. It will produce three output images files, a 32-bit floating point hue image, a unsigned 8-bit intensity image, and a 32-bit floating point saturation image. The following information is transferred from the input files to the output files(if appropriate):

Image labels
Mosaic labels
Projection labels
History file

Uses

An example use of this program is a contrast stretch applied in HIS space instead of RGB space. For example, first transform the RGB image to HIS images. Then apply a linear contrast stretch to the saturation image. Finally, transform the hue, intensity, and the enhanced saturation image back to RGB space.

Another example use of this program is data set merging. For example, given 3 Landsat TM images with 30 x 30 meter pixels and a registered SPOT panchromatic image with 10 x 10 meter pixels, the SPOT image can be merged with the Landsat images using this program. First transform the Landsat images from RGB to HIS space. Then replace the intensity image with the SPOT image. Finally, transform the hue, new intensity, and saturation images back to RGB space.

This procedure preserves the spectral information contained in the TM data while merging in the structural information contained in the SPOT image.

Examples

False color composite of Landsat TM Bands 2/3/4.

[Example Image]

From left to right, hue, intensity, and saturation components of above color composite.

[Example Image]