Perm National Research Polytechnic University, Berezniki branch (Russia):

Zatonskiy A. V., Doctor of Engineering Sciences, Professor, zxenon2000@yandex.ru

Perm National Research Polytechnic University, Berezniki branch (Russia):
Varlamova S. A., Ph. D. in Engineering Sciences, Associate Professor, varlamovasa@mail.ru

The paper defines the problem of automatic recognition of froth parameters in potassium ores flotation. A flotation engineer visually appraises froth in flotation cell and periodically changes flotation parameters. The problem consists in the factor of possible human error in flotation process control. The existing systems of froth product recognition in flotation cells, as well as bubbles images recognition methods are analyzed. Their main drawback — high requirements to image quality, is shown, demanding special conditions for camera filming, and accessories, unacceptable in potassium ores flotation. In order to solve the problem of froth color determination, two algorithms for image color profile definition have been developed. Performability of each algorithm is proved. A hypothesis is put forward about a relation between bubbles interflares distance and half sum of bubbles’ diameters. Image flares enhancing algorithm has been developed. Image binarization methods were analyzed, and a most effective one was chosen. Flare center determination algorithm is proposed. Software has been developed for statistical distribution of bubble sizes and mineral disseminations upon froth surface. It was checked through comparison of the software calculation results in four randomly chosen low-quality photographs with hand bubbles calculation results in the photograph. A close agreement of the results is demonstrated. In summary, over time duration within 0.8 s on a photograph of 72 dpi quality, the developed software creates statistical distribution of bubble sizes with a probability of 75 %, with an error within 10 %. This permits to use the described method for automatic recognition of froth parameters in flotation process automatic control or with a view to support decisions taken by a flotation engineer.

1. Flotatsionnyye tekhnologii Outotec (Outotec Flotation Technologies). URL: http://www.outotec.com/ImageVaultFiles/id_706/d_1/cf_2/OTE_Outotec_Flotation_technologies_rus_web.PDF (accessed date 18.03.2016).
2. Lavrinenko A. A. Modern flotation machines for mineral raw materials. Gornaya tehnika (Mining Engineering). Catalogue, year-book. St. Petersburg, Publishing House «Slavutich», 2008, pp. 186–195.
3. Forth M., Broussaud A., Monredon Th., Grebeneshnikov A. L., Kokorin A. M., Luchkov N. V., Smirnov A. O. The new generation of Metso Minerals flotation equipment — the foundation for effective solutions. Mining Industry Journal, 2005, No. 5 (60), pp. 21–24.
4. Tekhnologii flotatsii FLSmidth Dorr-Oliver Eimco (FLSmidth Dorr-Oliver Eimco Flotation Technologies). URL: http://www.flsmidth.com/~/media/Brochures/Brochures%20in%20RUSSIAN/FLSMinerals_RUSSIAN/FLSmidthDOEFlotationRUSemailv0.ashx (accessed date 17.03.2016).
5. Meshcheryakov N. F. Konditsioniruyushchie i flotatsionnyye apparaty (Conditioning and flotation devices). Moscow, Nedra, 1990, 237 pp.
6. Outotec FrothSenseTM. URL: http://www.outotec.com/ImageVaultFiles/id_735/d_1/cf_2/OTE_Outotec_FrothSense_eng_web.PDF (accessed date 17.03.2016).
7. Mashevskyi G. N., Petrov A. V., Moilanen J., Timperi J., Saloheimo K., Kemppinen H. Principles of computerised control of flotation process based on new Outotec product — FrothMaster videosystem. Tsvetnye Metally = Non-ferrous Metals, 2010, No. 2, pp. 89–92.
8. Romanenko S. A., Olennikov A. S. FrothMaster video system introduction experience practice at the «Green Mountain–2» Plant. Obogashchenie Rud = Mineral Processing, 2014, No. 2, pp. 23–28.
9. Minerals Processing Solutions. Automation. URL: http://www.metso.com/miningandconstruction/MaTobox7.nsf/DocsByID/0AF54610F6C67DDBC2257B030037F030/$File/Automation%20ENGLISH%20Low%20Res.pdf (accessed date 17.03.2016).
10. Khasanov B. R., Pol’kin K. V. Flotation froth as a controlled object. Sbornik trudov Mezhdunarodnoy nauchno-prakticheskoy konferentsii «Uralskaya gornaya shkola — regionam» (Proc. of Intern. research and practical conf. «Ural Mining School for regions»). Ekaterinburg, Ural State Mining University, 2013, pp. 344–351.
11. Rozentsov A. A., Evdokimov A. O., Grigoryev A. V. Plain images recognition of group point objects under detection errors presence. Izvestiya Vysshikh Uchebnykh Zavedeniy. Priborostroenie = Journal of Instrument Engineering, 2006, Vol. 49, No. 4, pp. 59–65.
12. Batyukov S. V. Selection of image information signs recognition method for analysis of flotation froth. Information Technologies and Systems 2011 (ITS 2011). Proc. of Intern. Conf., Minsk, BSUIR, 2011, pp. 98–99.
13. Shurygin A. M. Pattern recognition with help of interpoint distances. Doklady Mathematics, 2005, Vol. 72, No. 3, pp. 959–961.

14. Ordowski М., Meyer G. G. L. Geometric linear discriminat analysis for pattern recognition. Pattern Recognition, 2004, Vol. 37, Iss. 3, pp. 421–428.
15. Yangi Z., Peng X. F., Chu C. P., Lee D. J. Image processing and geometric parameters extracted from sliced image of porous biomaterial. Advanced Powder Technology, 2007, Vol. 18, Iss. 2, pp. 187–213.
16. Ekinci M., Aykut M. Palmprint recognition by applying wavelet-based kernel PCA. Journal of Computer Science and Technology, 2008, Vol. 23, Iss. 5, pp. 851–861.
17. Shapiro L. G., Stockman G. C. Computer Vision. Upper Saddle River, NJ, Prentice-Hall, 2001, 617 pp.
18. Shi J., Liu X.-G. Product quality prediction by a neural soft-sensor based on MSA and PCA. International Journal of Automation and Computing, 2006, Vol. 3, No. 1, pp. 17–22.
19. Nikitin K. V., Benderskaya E. N. A new approach to the use of recurrent neural networks for pattern classification problem solution. St. Petersburg State Polytechnical University Journal, 2007, No. 52-2, pp. 85–92.
20. Zatonskiy A. V., Varlamova S. A. Recognition of bubbles edges at the potash ores froth flotation. Materialy II Mezhdunarodnoy nauchno-prakticheskoy konferentsii «Virtualnoye modelirovanie, prototipirovanie i promyshlennyy dizayn» (Proc. of II Intern. research and practical conf. «Virtual simulation, prototyping and industrial design»). Tambov, Tambov State Technical University, 2015, pp. 220–223.
21. Sezgin M., Sankur B. Survey over image thresholding techniques and quantitative performance evaluation. Journal of Electronic Imaging, 2004, Vol. 13 (1), pp. 146–168.
22. Varlamova S. A., Myasnikov A. A. Choice of binarization method for recognition of foam parameters. New University. Technical Sciences, 2015, No. 5–6, pp. 73–78.
23. Zatonskiy A. V., Varlamova S. A. Perspective algorithm for recognition of foam parameters in the chemical-technological processes. Modeling and Information Technologies: Selected Papers of the International Scientific School «Paradigma» (Summer– 2015, Varna, Bulgaria). Yelm, WA, USA, Science Book Publishing House, 2015, pp. 52–56.