The recent progress of computer-based technologies, the speed-up of calculation instruments, and the decrease in hardware costs place at the geologist’s disposal powerful research tools. Such tools are particularly appreciable for measuring tasks, in that they allow the researcher to avoid repetitive and time-consuming operations. Therefore, the researcher can consider a larger number of cases that in the past, with a consequent increase in the reliability of results.
One of the application fields that are well suited to computer-based automation is the field of morphometric studies of clasts. By using image processing techniques, it is possible to process digital images of clasts and to extract the morphometric parameters of a large number of samples in an automatic way. A quantitative evaluation of such features for a certain number of clasts allows one to make objective comparisons among sediments of different origin and "sedimentary history".
The method is developed in three steps:
• photographic acquisition (with an optical microscope) of quartz grains (600<diameter<210micron) and digitalization of the enlarged photographs by means of a scanner;
• image pre-processing (reduction of acquisition noise, conversion to gray levels and segmentation);
• feature extraction by an automatic method expressly devised and validated.
