@unibiblio

Image processing code for characterization of multiphase flow in porous media

, and . Software, (2024)Related to: Vahid Dastjerdi, S.; Karadimitriou, N.; Hassanizadeh, S. M. & Steeb, H.: Experimental evaluation of fluid connectivity in two-phase flow in porous media during drainage. Water Resources Research 58 (2022), e2022WR033451. doi: 10.1029/2022WR033451.
DOI: 10.18419/darus-4153

Abstract

This work utilizes microfluidic experiments to gather data captured as snapshots during the experiments. These snapshots provide real-time information and undergo image processing to derive the required data. Image processing involves several steps tailored to the investigations: Making a reference image (mask): This process involves creating a reference image, or mask, to document the initial conditions. For instance, the porous domain is imaged when saturated with one phase to differentiate various areas containing different phases. Reading and cutting images: Images showing changes in fluid volume fraction are selectively chosen and processed. Each image is read into MATLAB, and the area of interest is extracted. Image segmentation: Labeling each pixel of the images is done via thresholding and edge detection. Measuring parameters: Parameters like saturation, interfacial length, area, contact angle, and curvature are measured. These parameters play a crucial role in analyzing the experiments. The interfacial area is calculated through various formulations. Calculating capillary pressure: Several forms of capillary pressure are calculated using information derived from the experiments. REV-Scale Quantities: Parameters are upscaled to represent Representative Elementary Volume (REV)-scale values essential for continuum theories. REV-scale capillary pressure is derived from pore-scale values using appropriate averaging techniques.

Links and resources

Tags