Material identification using multiple images

Abstract

A computer-implemented method of image processing for materials analysis is provided. At least three image datasets are obtained, these representing intensity values of image pixels and being in common spatial registration. The image datasets are processed so as to assign a comparison measure to each pair of image datasets, the comparison measure for a given pair of image datasets being representative of the difference between the spatial intensity information within the pair. A number of image datasets are then selected using the comparison measures. A color difference measure is defined which represents the difference between pairs of colors of a color set. Colors are assigned to the selected image datasets such that pairs of the selected image datasets which have substantially different spatial intensity information are assigned respective colors which have a substantially different color difference measure. A number of the selected image datasets are then combined together to form an output color image dataset for the formation of a color image. The combination is such that each pixel of the color image takes on a color according to the relative intensities and colors of the selected image datasets so that if the pixel intensity of one dataset is substantially greater than the sum of the pixel intensities for all the other selected datasets, then the output color in the respective part of the image will substantially match the color assigned to that image dataset.