ALGORITHM FOR STATISTICAL ANALYSIS OF THERMAL IMAGING DATA FOR ASSESSING THE STRUCTURE OF THE OCCUPATION LAYER OF ARCHAEOLOGICAL SITES
A. G. Zlobina1, I. V. Zhurbin1, A. S. Shaura1, E. A. Rubleva1,2
1Udmurt Federal Research Center, Ural Branch, Russian Academy of Sciences, Izhevsk, Russia
2Udmurt State University, Izhevsk, Russia
Keywords: Haralick texture features, image segmentation, thermal imaging, archaeological site, reference data, boundary
Abstract
The use of thermal imaging for reconstructing historical heritage sites is limited by the influence of modern plowing and the absence of archaeological sites with contrasting thermal properties. A distinctive feature of the proposed statistical analysis algorithm is the use of Haralick texture features at the stage of feature extraction. Subsequent segmentation of the image makes it possible to identify sections of the occupation layer of various capacities as areal landscape objects. The interpretation of their properties is based on the reference data of interdisciplinary research - excavations, geophysics, and soil drilling. An assessment of the trends in the relative locations of segments of different classes and the ratio of their areas allows us to put forward a hypothesis about the boundaries of an archaeological site. At the site of the medieval Kushmanskoe III settlement (Udmurtia), the analysis of the structure of the segmented thermal image reveals fundamental differences in the distribution of the humus layer in the settlement and the adjacent territory, which was not subjected to anthropogenic influence. The proposed algorithm can be considered as an effective tool for studying historical heritage sites.
|
