Image Analysis of Size and Shape of Mineral Particles

Mineral analysis is very important in engineering geology. It is often that a geologist analyzes minerals on an image under microscope, or by taking photos. It is difficult to recognize minerals automatically due to the complexity of mineral images. This paper presents the newly studied an image segmentation algorithm for mineral particles. The algorithm consists of image pre-processing, gray level image segmentation, background split, merge of small regions, and merge and split based on shape analysis. The developed algorithm has been tested in a laboratory and works satisfactory.