Fast extraction of bubble shape parameters algorithm on GPU

Afef Bettaieb, Nabila Filali, Taoufik Filali, Habib Ben Aissia


Bubble shape and dimensions play a key role in the study of various gas/liquid two-phase flow systems. Various techniques based on the digital image processing for the study of bubble characteristics in gas/liquid two-phase flow were introduced. In this paper, a new approach for real-time parameters estimation of air bubbles is proposed. To extract with efficiency their contours, an edge detection algorithm based on the transformation of raw images in first step and contour thinning in second step is proposed in this study. An algorithm based on bounding box was applied to the contour image to extract the parameters of bubbles. Both algorithms are implemented on an nvidia GTX780 GPU. The extracted parameters of bubbles could provide the database for further study of the gas/liquid flow.

Total views : 10 times


  • There are currently no refbacks.

View IJAAS Stats

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.