A Machine Learning Approach to Tracking and Characterizing Planar or Near Planar Fluid Flow

A Machine Learning Approach to Tracking and Characterizing Planar or Near Planar Fluid Flow

Mahendra Gooroochurn, David Kerr, Kaddour Bouazza-Marouf
Copyright: © 2020 |Pages: 12
DOI: 10.4018/IJNCR.2020070105
(Individual Articles)
No Current Special Offers


This paper presents a framework to segment planar or near-planar fluid flow and uses artificial neural networks to characterize fluid flow by determining the rate of flow and source of the fluid, which can be applied in various areas (e.g., characterizing fluid flow in surface irrigation from aerial pictures, in leakage detection, and in surgical robotics for characterizing blood flow over an operative site). For the latter, the outcome enables to assess bleeding severity and find the source of the bleeding. Based on its importance in assessing injuries and from a medical perspective in directing the course of surgery, fluid flow assessment is deemed to be a desirable addition to a surgical robot's capabilities. The results from tests on fluid flows generated from a test rig show that the proposed methods can contribute to an automated characterization of fluid flow, which in the presence of several fluid flow sources can be achieved by tracking the flows, determining the locations of the sources and their relative severities, with execution times suitable for real-time operation.
Article Preview

Fluid flow characterization has applications in numerous fields, especially where different phases of fluid are present. For example, Schubert (2016) designed an experimental set up to generate signature patterns of fluid flow on distillation trays with the aim to improve the fluid flow and achieve better efficiency. A wire-mesh sensor technique is used to model the flow pattern. Similarly Severin (2017) proposes a minimal invasive sensor solution for parallel salinity measurement aimed at characterizing the hydrodynamics of reactors handling aqueous liquid. Wen (2018) developed an image processing technique to measure volume and surface area of bubbles, proposing an algorithm suitable to characterize large and severely deformed bubbles, making it possible to relate the volume of the bubbles to the gas flow rate. Mahmoodi (2018) demonstrates the efficacy of the image processing algorithms in LabVIEW® for automated advanced image processing to observe fluid transport mechanisms and characterize fluid flow. Johansen (2010) used high speed tomographic imaging to achieve frame rates of several thousand images per second, thus overcoming blurring and inconsistent measurement obtained with fourth generation X-ray systems for characterizing the dynamics in multiphase flows. In vision-based systems, it is common to first enhance the image by using appropriate image processing operations before features can be reliably extracted. For example, morphological operations have been applied by Bhateja (2019) and Nigam (2020) to enhance low contrast and noisy MRI images for brain tumor detection, while Bao (2017) developed a framework for optimizing feature selection from video streams using a machine learning model.

Complete Article List

Search this Journal:
Volume 12: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 11: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 2 Issues (2017)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing