论文部分内容阅读
土图像的数字特征是土状态的重要反映。本文使用不同土层的扫描电镜(SEM)图像,在确定土颗粒与孔隙分割阈值的基础上,得到了土粒的圆形度、矩形度等形状特征参数,计算了基于灰度直方图和灰度共生矩阵的纹理特征参数,分析了这些微观参数与宏观工程性质指标之间的关系。结果表明,不同土层中,土粒多呈扁圆状、离散性较大,黏土内切圆半径较小;黏土和淤泥质黏土层的图像对比度比粉质黏土和砂质粉土层的要大;黏性土的孔隙比越大,土颗粒的圆形度和矩形度就越大,SEM图像的对比度、标准差、熵也就越大,土粒的形状离散指标和偏心率则越小;黏性土的黏聚力越大,土粒的圆形度和矩形度就越小,而SEM图像的形状离散指标和偏心率就越大。
The digital character of the soil image is an important reflection of the state of the earth. In this paper, the scanning electron microscope (SEM) images of different soil layers were used to determine the shape parameters of soil particles, such as circularity and rectangularity. Based on the gray histogram and gray Degree co-occurrence matrix texture feature parameters, the relationship between these microscopic parameters and macroscopic engineering properties index is analyzed. The results show that in different soil layers, the soil particles are mostly round and flat with large discreteness and small radius of the inscribed circle in the clay. The image contrast of the clay and silty clay layers is lower than that of silty clay and silty clay layers The larger the clay ratio is, the larger the circularity and the squareness of the soil particles are. The larger the contrast, the standard deviation and the entropy of the SEM image, the smaller the dispersion index and the eccentricity of the soil particles ; The cohesion of cohesive soil is larger, the smaller the circularity and the squareness of soil particles, the larger the dispersion index and the eccentricity of the shape of SEM image.