Tracking the Dispersed Phase in Solid–Liquid Two-Phase Flow Using Digital X-Ray Imaging and the CSRT Tracking Algorithm in Python

Sharifzadeh, Mohsen; Askari, Mojtaba; Beigzadeh, Amirmohammad; Ardini, Hadi

doi:10.22078/jpst.2025.5817.1993

Tracking the Dispersed Phase in Solid–Liquid Two-Phase Flow Using Digital X-Ray Imaging and the CSRT Tracking Algorithm in Python

Articles in Press

Document Type : Research Paper

Authors

Mohsen Sharifzadeh ¹

Mojtaba Askari ²

Amirmohammad Beigzadeh ²

Hadi Ardini ²

¹ Radiation Application Research School, Nuclear Science and Technology Research Institute, Tehran, Iran

² Radiation Application Research School, Nuclear Science and Technology Research Institute, Tehran, Iran.

10.22078/jpst.2025.5817.1993

Abstract

The dispersed-phase velocity measurement module in two-phase flow represents one of the most critical components of a two-phase flowmeter, with particular importance in monitoring oil, gas, and pipeline-transported products. Compared with conventional velocimetry methods, digital X-ray imaging offers a novel solution due to its ability to penetrate opaque materials, its high sensitivity to density variations, and its capability for rapid imaging of dynamic processes. By employing advanced image-processing algorithms, the dispersed phase can be identified and tracked effectively. This study investigates the feasibility of tracking and computing the trajectory equation of the dispersed phase through the integration of digital X-ray imaging and the CSRT (Discriminative Correlation Filter with Channel and Spatial Reliability) tracking algorithm. The innovation of this research lies in applying image-processing algorithms to compute particle trajectories in multiphase flow metering systems. Implemented in Python, the proposed method achieved particle detection and tracking accuracy of 94.7% with an error below 3% under laboratory conditions. Results demonstrated that combining digital X-ray imaging with the CSRT tracking algorithm enables trajectory recognition of dispersed particles with a maximum error of ±5%.

Keywords

Tracking

Dispersed Phase

Two-Phase Flow

X-Ray

CSRT Algorithm