Image processing for bubble morphology characteristics in diluted bubble swarms
Résumé
The present investigation describes, in detail, a data processing strategy for characterizing bubble sizes and shapes properly from shadow graph images in diluted bubble swarm conditions. A flat pseudo-two-dimensional bubble column is studied, where two different bubble types (ellipsoidal and spherical caps) can be injected through two different spargers. Various statistical equivalent diameters (D-10, D-20, D-30, D-32) are considered, and the morphology is characterized via the eccentricity (x) as a function of the size. To further broaden the experimental conditions, several Newtonian liquids characterized by different viscosities and gas injection rates ranging from 50 to 200 l/h, are used. Additionally, an original method for verifying results, using control parameters such as orientation and solidity, is presented.