*Nobuo Geshi1, Rina Noguchi2, Motomaro Shirao, Daigo Shoji3, Fujimoto Keiichiro3, Kodai Ikeya4, Junichi Haruyama3
(1.Geological Survey of Japan, The National Institute of Advanced Industrial Science and Technology, 2.Faculty of Science, Niigata University, 3.JAXA, 4.Tokai University)
Keywords:field geology, outcrop, image analysis
Visual observation of the outcrop is the first step of field geology. Generally, geologists observe the outcrop to grip the overall structure and components of the outcrop, then they do the various instrumental measurements and samplings for laboratory analysis for the interpretation of the geoprocess that forms the geological unit exposing the outcrop. Thus, the analysis of the optical image information of the outcrop is the fundamental process of field geology. The visual characteristics of the geological units depend on their mechanical properties, meso-micro scale internal structures, and components. Geologists "read" the optical information and make their preliminary judgments about the structure and components of the outcrop, then they will make close-up investigations to follow up the judgments. However, this process is generally done unconsciously and rarely clearly specified. This may make a difficulty of the training of geological field work at the primary level of education, and the automation of the identification of geological structure. Here, we show the examples of the identification process of the contents of geological units and their boundaries using some representative subaerial outcrops in the basaltic volcanoes. We chose some representative outcrops of Izu-Oshima and Miyakejima consisting of air fall deposits, pyroclastic flow deposits, lava flows, and secondary deposits for analysis. Then, we tried to build an identification flow to distinguish air fall deposits (scoria fall, ash fall), lava flow (central massive part and clinker pats), pyroclastic flow deposit (non-welded, massive, or stratified), and secondary aerial deposit, with step-by-step identifications of each characteristic. Using some representative outcrops in volcanic fields, we analyze the identification process of the geological structures exposed in the outcrop based on their visual characteristics. The identification process of the visual information of the outcrop is expected to be applied for the automatic analysis of the optical images of the outcrops.