SOIL STRUCTURE EVALUATION BASED ON THE FRACTAL THEORY AND IMAGE PROCESSING TECHNOLOGY

Fractal thcory and image processing technology has been more and more applied to the soil structure evaluation recent years. In this study, dredger fill from the reclamation area of Chongming Island, Shanghai, were tested for the soil structure by sieve, densimeter, LPSA, MIP and SEM. Moreover, fractal values of the particles and pores based on the fractal theory, as well as the morphological and statistical parameters based on computer image processing technology were both obtained; correlations between fractals with the morphology and the discreteness of directional angle were further analyzed. The results revealed that the fractal dimension of particle size distribution (PSD) had two scale -free intervals (SF1) with 0.075 mm as an important demarcation point. The 'Wenger model was more suitable to describe the fractal characteristics of the relatively smaller pores (< 4 um). An abnormal segment caused by the inkbottle pores occurred at the junction of the macropore and the mesopore. Fractal values obtained from SEM had some limitations in describing the real complexity of the soil particles and pores. Pores with larger equivalent diameter tended to have more complex structures; the directional angle discreteness of the soil particle had some influence on its complexity.

Automated summary

Fractal theory and image processing technology have been increasingly used for soil structure evaluation. The study found that particle size distribution had two scale-free intervals, and the fractal characteristics of pore structure could be described using the "Wenger model".