MEASUREMENT OF GEOMETRICAL PARAMETERS OF THE CRUDE-OIL/WATER INTERFACE PROPAGATING IN MICROFLUIDIC CHANNELS USING DEEP LEARNING TOOLS

• DOI: 10.1615/interfacphenomheattransfer.2022045682
• Publication Date: 2022-12-06T19:06:41Z
• Authors: Fabio Grazioso; Viktor M. Fliagin; Natalia A. Ivanova
• Keywords: Meniscus; Interface (matter)

This paper reports the results of application some software tools based on deep learning models, processing microscopic images interface between crude oil and water, while propagating in microfluidic channels. The U-Net model is used to classify pixels separate them from rest (semantic segmentation). has allowed for automatic measurement geometric parameters meniscus, making possible large amounts images. Live videos meniscus have been recorded water propagates guides previously filled with oil, then frames (images) video extracted processed. In this way, we were able consider information about time also study dynamic behavior parameters. Among that it measure, angle walls propagation channel chosen. measured was compared contact a static regime, method sessile drop.