Practical object and flow structure segmentation using artificial intelligence

Journal article (2024)

Authors

AR Khojasteh Fluid Mechanics - Mechanical, Maritime and Materials Engineering
W. van de Water Fluid Mechanics - Mechanical, Maritime and Materials Engineering
J. Westerweel Fluid Mechanics - Mechanical, Maritime and Materials Engineering
Research Group
Fluid Mechanics (Mechanical, Maritime and Materials Engineering) (TU Delft)
More Info
expand_more

Published Date

2024

Language

English

Faculty

Mechanical, Maritime and Materials Engineering

Department

Process and Energy

Research Group

Fluid Mechanics

Abstract

This paper explores integrating artificial intelligence (AI) segmentation models, particularly the Segment Anything Model (SAM), into fluid mechanics experiments. SAM’s architecture, comprising an image encoder, prompt encoder, and mask decoder, is investigated for its application in detecting and segmenting objects and flow structures. Additionally, we explore the integration of natural language prompts, such as BERT, to enhance SAM’s performance in segmenting specific objects. Through case studies, we found that SAM is robust in object detection in fluid experiments. However, segmentations related to flow properties, such as scalar turbulence and bubbly flows, require fine-tuning. To facilitate the application, we have established a repository (https://github.com/AliRKhojasteh/Flow_segmentation) where models and usage examples can be accessed.