La descripción de las trazas de uso en materias primas volcánicas: resultados preliminares de un análisis estadístico descriptivo

Abstract

Most studies of use-wear analysis are explanations of alterations due to use observed on siliceous rocks of sedimentary origin –principally flint. Studies of other raw materials have been left aside. Therefore, most explanations of use wear are based on a single source of data, as if all rocks behave exactly in the same way. Studying obsidian, metamorphic rocks, quartzite, basalt, etc. we have found a lack of any references to explain use wear on these raw materials.

This problem is more relevant when using digital image processing techniques to study alterations by use. This method is based on the analysis of luminance values among image pixels, which describe the surface textures. We should find the way of representing alterations by use and their features through a language based on pixel values.

We propose the analysis and evaluation of several intervals of light spectrums. Alterations by use on flint surfaces (what are usually called micropolished areas) appear in an image on the highest brightness of the luminance interval. On volcanic rock surfaces, alterations from use appear at different light intervals. The dark band of the spectrum should be taken into account when studying non-flint raw materials.

We have developed an experimental program using three varieties of andesite and one of obsidian. With tools made of these raw materials, we have reproduced three different tasks: cutting shell, scraping wood and scraping dry fur. In each case, we have obtained five digital images which showed alterations due to use. Each image was segmented using different light intervals through a density slice (0-80, 0-120 and 160-200 from the standard grey scale of 0-255). The segmentation process generates restricted areas, which are used as the basic elements of use-wear description. These are the texture elements (or texels), which are statistically processed to calculate differences and recurrences between them.

We characterise use-wear image features using the same methodology applied in the chapter 6 of this volume. Both papers share the same goal: a quantified description of lithic use wear for developing knowledge that allows us to classify the features of the alterations by use according to their discriminating power. In the present paper, we also try the application of the same methodology to a more complex problem, which is a different phenomenological development of the alterations by use.