Pattern Recognition by DTW and Series Data Mining in 3D Stratum Modelling and 3D Visualization

Abstract

Three-dimension (3D) modeling and visualization of stratum plays important role in seismic active fault detection, of course in geoinformation science. Well-logging data of strata is taken as time series. Similarity measure of subsequence search is proposed based on dynamic time warping (DTW). Realizing time series match in different length of time series. The frequent pattern mining experiment is carried on to survey data of strata by multivariable combination analysis. We supply the stratum geophysics attribute by the depth (time) records sequence in the non-drill hole survey data's places using these frequent patterns,combining structure frame and mathematics geology interpolation technology, establish 3D geology model of the target area,and develop the underground geologic body 3D visualization software depending on visual studio.net and OpenGL graph packages, realize 3D visualization system.