THE APPLICATION OF SVM AND FMI TO THE LITHOLOGIC IDENTIFICATION OF VOLCANIC ROCKS
ZHANG Ying1, PAN Bao-zhi2
1. College of Information, Guangdong Ocean University, Zhanjiang 524088, China;
2. College of Geoexploration Science and Technology, Jilin University, Changchun 130026, China
From the viewpoint of chemical composition categorization and structure classification of rocks, an effective method was proposed to identify the lithology of volcanic rocks by using logging data. On the one hand, the conventional logging data could be obtained by core wafer identification. Thus, after processing the data with Support Vector Machines (SVM) method of statistical theory, we could get the lithologic type of the volcanic rocks, which are classified according to the chemical composition of rocks. On the other hand, the volcanic rocks can be classified as volcanic lava, pyroclastic lava and pyroclastic rock according to the rock structure. Typical formation micro-resistivity imaging logging (FMI) image mode can be concluded by establishing the corresponding relationship between FMI images and lithology of volcanic rocks with different structures. As a result, the lithologic type of the volcanic rock classified by rock structure can be determined. Finally, by combining these two kinds of lithology, the ultimate rock lithology can be determined, too. In this paper, the authors presented a novel method to identify the lithology of volcanic rocks by combining SVM processed logging data and FMI image mode.
张莹, 潘保芝. 支持向量机与微电阻率成像测井识别火山岩岩性[J]. 物探与化探, 2011, 35(5): 634-638,642.
ZHANG Ying, PAN Bao-zhi. THE APPLICATION OF SVM AND FMI TO THE LITHOLOGIC IDENTIFICATION OF VOLCANIC ROCKS. Geophysical and Geochemical Exploration, 2011, 35(5): 634-638,642.