Deep Learning Bubble Segmentation on a Shoestring

Abstract

Image segmentation in bubble plumes is notoriously difficult, with individual bubbles having ill-defined shapes overlapping each other in images. In this paper, we present a cheap and robust segmentation procedure to identify bubbles from bubble swarm images. This is done in three steps. First, individual, nonoverlapping bubbles are detected and isolated from true experimental images. In the second step, these bubble images are combined to generate synthetic ground truth images. In the third and final step, the synthetic images are used as training data for a machine learning script. The trained model can now be used to segment the data of experimental bubble swarms. The segmentation procedure is demonstrated on three different experimental data sets, and general observations are discussed.

Multiscale Modeling and Artificial Intelligence for Multiphase Flow Science

“…AI was also employed for image analysis to tackle the challenging task of segmenting bubbles. Homan et al developed a cost-effective and reliable segmentation method to identify bubbles from bubble swarm images; the model successfully segmented experimental bubble swarm data, demonstrating its effectiveness. Jin et al developed a machine learning-assisted image segmentation method for automatic bubble identification in gas–solid quasi-2D fluidized beds, achieving high accuracy with minimal training data.…”