A proximity-based image-processing algorithm for colloid assignment in segmented multiphase flow datasets

Area of Science:

• Environmental Science
• Engineering
• Geoscience

Background:

• Colloidal transport and deposition are critical in environmental and engineering contexts.
• X-ray microtomography (XMT) offers advanced visualization capabilities for these processes.
• Limited data exists on colloid segmentation and spatial partitioning in porous media.

Purpose of the Study:

• To develop and demonstrate an image processing approach for segmenting and classifying colloidal particles based on their location within porous media.
• To enable quantitative analysis of colloid spatial partitioning relative to different phases (e.g., fluid interfaces).

Main Methods:

• Utilized X-ray microtomography (XMT) for high-resolution imaging of porous media.
• Developed an algorithm to segment colloidal particles and aggregates.
• Implemented a phase-based classification method using distinct markers for attachment sites.

Main Results:

• Successfully demonstrated the image processing algorithm's efficacy on an XMT dataset from a drainage experiment.
• Enabled comparison of flow conditions, fluid, and colloid properties with observed colloid behavior.
• Provided a framework for elucidating colloidal deposition mechanisms and attachment site importance.

Conclusions:

• The presented algorithm effectively assigns segmented colloids to different partitioning classes.
• This method enhances understanding of colloidal deposition, fluid property influences, and colloid morphology in porous media.
• Facilitates detailed analysis of colloid behavior in saturated and unsaturated conditions.