GIS-based Quantitative Analysis for Mesostructure of Cohesive Soil Containing Calcareous Nodules

摘要

A spatial statistic method for quantitative analysis of cohesive soil containing calcareous nodules' mesostructure was illustrated in ArcGIS. Through cluster unsupervised classification methods of raster image processing, calcareous nodules can be extracted from soil mesostructure images. Using standard deviational ellipse, the orientation angles of calcareous nodules can be fast determined, and the area, perimeter, thickness and characteristics of ellipse of calcareous nodules can be measured by geometry statistics. Based on the above statistical data, the granularity, eccentricity and orientation distribution can be obtained by fractal analysis. The results show that the method is more scientific, reliable and efficient for the quantitative analysis of cohesive soil containing calcareous nodules' mesostructure.