MINERAL RECOGNITION IN DIGITAL IMAGES OF ROCKS - A NEW APPROACH USINGMULTICHANNEL CLASSIFICATION

Digital image-analysis is used for counting (modal analysis) and measu ring grains (texture and fabric analysis). Traditionally, minerals are visually recognized and manually outlined prior to digitizing and sub sequent analysis. This limitation can be overcome by using multichanne l methods of classification, in which the minerals in multichannel dig ital images are accurately recognized on the basis of their unique spe ctral or elemental signatures, established by a training stage prior t o classification. The technique is applied here for the modal analysis of three-color, electro-optical images digitized using CCD video came ras and scanners, and multi-elemental X-ray maps acquired with an elec tron microprobe. In all three case studies of the analysis of plutonic igneous rocks, the resulting mineral modes are sufficiently precise t o identify significant compositional heterogeneities between groups of samples and, in the case of X-ray element maps, to quantify the degre e and type of alteration.