Foam Identification Based on Morphological Algorithm and Hough Transform

Abstract:

Petrochemical pharmaceutical bio-fermentation industry, witch is characterized by the production press of the material inside the device sets experiencing recessive chemical processes. In industrial production process, some media usually easy to produce large foam, witch affect production, it must be detected the amount of foam in real time, to effectively eliminate the foam bubbles and control the processes of the physical and chemical reactions to ensure the efficient, orderly and safe conduct throughout the process .Foam is multi-interface, uneven density and unstable surface shape, each bubble has its own gas and liquid interface, many bubbles overlap together, with the various interfaces. Staggered and different angles, witch making it light-scattering, in order to obtain a clear picture of foam, and effectively obtain features of foam, shoot on goal from different angles, the captured images are compared and analyzed to choose the best camera angle. Morphological methods was introduced and used to extract the edge of bubbles in foam. This work proposed a improved Hough Transform (HT) to analyze round bubbles and detect each bubble.To detect whether the captured image is the foam image, this work also proposed calculating the density of bubbles in the captured image. The preliminary results showed that the algorithm is dynamic with high efficient.

• Periodical: Applied Mechanics and Materials (Volumes 602-605)
• Pages: 1745-1748
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.602-605.1745
• Online since: August 2014
• Authors: Le Kang*
• Keywords: Hough Transform (HT), Image Processing, Morphological Algorithm, Parameter Extraction

References

[1] Kuo W.F., Lin C.Y., Sun Y.N. Region similarity relationship between watershed and penalized fuzzy Hopfield network algorithms for brain image segmentation. Inernational Journal of Patern Recognition and Artificial Intelligence, 2008, 22(7): 1403-1425.

DOI: 10.1142/s0218001408006788

[2] Mignotte M. Segmentation by fusion of histogram-based K-means clusters in different color spaces. IEEE Transactions on Image Processing，2008, 17(5): 780-787.

DOI: 10.1109/tip.2008.920761

[3] Fei Z.G. Guo J.J., Wan P., et al. Fast automatic image segmentation based on bayesian decision-making theory. In: Proceedings of the Internaional Conference on Information and Automation, Zhuhai, Macau, 2009. 184-188.

DOI: 10.1109/icinfa.2009.5204917

[4] Faruquzzaman A.B.M., Paiker N.R., Arafat J., et al. Object segmentation based on spli and merge algorithm. In: Proceedings of the IEEE Region 10 Conference, Hyderabad, India, 2008. 1447-1450.

DOI: 10.1109/tencon.2008.4766802

[5] Lezoray O., Charrier C. Color image segmentation using morphological clustering and fusion with automatic scale selecton. Pattem. Recognition Letters, 2009, 30(4): 397-406.

DOI: 10.1016/j.patrec.2008.11.005