Quantifying Porosity through Automated Image Collection and Batch Image Processing: Case Study of Three Carbonates and an Aragonite Cemented Sandstone

Modern scanning electron microscopes often include software that allows for the possibility of obtaining large format high-resolution image montages over areas of several square centimeters. Such montages are typically automatically acquired and stitched, comprising many thousand individual tiled images. Images, collected over a regular grid pattern, are a rich source of information on factors such as variability in porosity and distribution of mineral phases, but can be hard to visually interpret. Additional quantitative data can be accessed through the application of image analysis. We use backscattered electron (BSE) images, collected from polished thin sections of two limestone samples from the Cretaceous of Brazil, a Carboniferous limestone from Scotland, and a carbonate cemented sandstone from Northern Ireland, with up to 25,000 tiles per image, collecting numerical quantitative data on the distribution of porosity. Images were automatically collected using the FEI software Maps, batch processed by image analysis (through ImageJ), with results plotted on 2D contour plots with MATLAB. These plots numerically and visually clearly express the collected porosity data in an easily accessible form, and have application for the display of other data such as pore size, shape, grain size/shape, orientation and mineral distribution, as well as being of relevance to sandstone, mudrock and other porous media.