基于CT数字图像的土壤孔隙分形特征研究

通过CT扫描获得了河南封丘地区3种不同质地潮土样本的高精度图像,利用数字图像分析方法识别出土壤孔隙及其轮廓线,然后用计盒法确定了表征土壤孔隙形态不规则性的分形维数.在此基础上,通过不同形式的分形模型预测了土壤水分特征曲线.结果表明,河南封丘地区不同质地潮土的孔隙分形维数(包括面积和轮廓线分维)差异不明显;预测水分特征曲线时,针对不同质地的土壤选择适宜的分形模型才能得到较好的模拟结果.

Characterizing Fractal Characteristics of Soil Pores Based on High-Resolution Digital CT Images

In this study, high-resolution digital CT image of different textural Fluvo-aquic soils in Fengqiu region of Henan Province obtained by computerized tomography. The pores and outlines of these images were recognized by image analysis technique. The fractal dimensions which described the fractal characteristics of the area and outlines of soil pores were deternined by box-counting technique. The soil water retention characteristics were predicted from the fractal dimensions using different models. Results indicated that there was no substantial difference in the area and outline fractal dimensions of pores for different textural Fluvo-aquic soils in Fengqiu region of Henan Province. A suitable fractal water retention model should be selected in order to get the more reliable simulated result.