On Image Processing Algorithm for Registration of Three-Dimensional Gas Bubble Movement Using a High Speed Camera

Abstract

The paper deals with an imaging computer tomography method based on simple image processing techniques for two phase flow analysis. Moreover, it has been presented the algorithm of 3D bubble trajectory reconstruction using a single high speed camera and the system of mirrors. In the experiment a glass tank filled with distilled water was used. The nozzle through which the bubbles were generated was placed in the center of the tank bottom. Through the use of basic image processing and analysis techniques such as noise reduction, smoothing, edge detection and few algorithms like close contour filling, tracking single bubble etc. proposed in paper it became possible to draw out 3D trajectories for gas bubble paths in liquid. In the paper the measurement error of imaging computer tomography method has been estimated. The maximum measurement error recorded for this method was within the limits ±0,65 [mm] for a certain set of parameters like: resolution, mirror angle and deviation error of z axis from the 90^o vertical line. Trajectories of subsequently departing bubbles were visualized in the form of figures.