Automatic mineral identification using evolutionary computation technology
is discussed. Thin sections of mineral samples are photographed digitally u
sing a computer-controlled rotating polarizer stage on a petrographic micro
scope. A suite of image processing functions is applied to the images. Filt
ered image data for identified mineral grains is then selected for use as t
raining data for a genetic programming system, which automatically synthesi
zes computer programs that identify these grains. The evolved programs use
a decision-tree structure that compares the mineral image values with one o
ther, resulting in a thresholding analysis of the multi-dimensional colour
and textural space of the mineral images.