Application of Digital Image Processing Technique in Geotechnical Dynamic Triaxial Test

Abstract

In soil studies, triaxial tests are commonly conducted to measure and analyze the deformations and strength characteristics of the soils under different stress paths. Since conventional triaxial test apparatus uses displacement sensors to measure soil deformations, it is difficult to obtain the local deformation of the soil and to capture the progressive failure of soils destruction process and morphology of the soils. The present paper applies digital image correlations measurement technique to measure the surface deformation of soil samples in static and dynamic triaxial tests. As a result, the non-contact surface displacement field is continuously monitored. The sub-pixel corner detection, unit node equation, distortion correction and error correction of object distance measurement are adopted to ensure the measurement accuracy. In order to ensure the stability of the illumination environment, the commonly used high lumen light source is replaced by the LED cold light source in the pressure chamber. The test results show that the dynamic deformation measurement method involving the use of digital image processing technology is effective in measuring the characteristics of the axial and radial deformation of the strain fields, thereby provides a strong experimental foundation to study the mechanical characteristics of soil samples.