Full text of article: Russian language.

Abstract:
We discuss a machine vision system (MVS) for determining the number of gel particles in a polymer solution when conducting the laboratory analysis. Components and specifications of the system intended to solve the problem are proposed. 
As part of the MVS development process, we come forward with novel methods for binary image thresholding and analysis, which prove to be efficient for low contrast images and at fairly high crosstalk levels. The software to realize the aforesaid methods has been developed, allowing the number of gel particles in the polymer solution to be automatically determined.
The use of the MVS instead of a human observer enables the psychovisual load on the laboratory staff to be essentially reduced, the accuracy and reliability of determining the number of gel particles in the polymer solution to be enhanced, as well as providing the automatic recording of the laboratory analysis data.

Key words:
machine vision system, image processing, image binarization, contour discontinuity identification.

References:

1. Bulanov, А. P. Machine vision system for recording tank railway cars / A. P. Bulanov, S. G. Volotovskiy, N. L. Kazanskiy, S. B. Popov, R. V. Khmelev, S. M. Shumakov, the journal of Industrial Automation, 2005. – # 6. – pp. 57-59 (in Russian).
2. Volotovskiy, S. G. Machine vision system for recognition of railway tank marking numbers with use of a modified correlator in the Hausdorff metric / S. G. Volotovskiy, N. L. Kazanskiy, S. B. Popov, R. V. Khmelev, Computer Optics, 2005. – # 27. – pp. 177-184 (in Russian).
3. Methods of Computer Optics (second, revised edition) / Ed. by V. A. Soifer - Moscow: Fizmatlit Publishers, 2003. – 784 pages (in Russian).
4. Sezgin, M. Survey over image thresholding techniques and quantitative performance evaluation / M. Sezgin and B. Sankur  // Journal of Electronic Imaging, 2004. – Vol. 13(1), – pp. 146–165.
5. Otsu, N. A threshold selection method from gray level histograms // IEEE Trans. Syst. Man Cybern., 1979. – SMC-9. – pp. 62–66.
6. Kittler, J. Minimum error thresholding / J. Kittler and J. Illingworth // Pattern Recogn., 1986. – Vol. 19. – pp. 41–47.
7. Kittler, J. On threshold selection using clustering criteria / J. Kittler and J. Illingworth // IEEE Trans. Syst. Man Cybern., 1985. – SMC-15. – p. 652–655.
8. Nakagawa, Y. Some experiments on variable thresholding / Y. Nakagawa and A. Rosenfeld // Pattern Recogn., 1979. – Vol. 11(3). – p. 191–204.
9. Deravi, F. Grey level thresholding using second-order statistics / F. Deravi and S. K. Pal // Pattern Recogn. Lett., 1983. – Vol. 1. – p. 417–422.
10. Sauvola, J. Adaptive document image binarization / J. Sauvola and M. Pietaksinen // Pattern Recogn., 2000. – Vol. 33. – p. 225–236.
11. White, J. M. Image thresholding for optical character recognition and other applications requiring character image extraction / J. M. White and G. D. Rohrer // IBM J. Res. Dev., 1983. – Vol. 27(4). – p. 400–411.
12. Bernsen, J. Dynamic thresholding of gray level images // ICPR’86: Proc. Intl. Conf. Patt. Recog., 1986. – pp. 1251-1255.
13. Yanowitz, S. D. A new method for image segmentation / S. D. Yanowitz and A. M. Bruckstein // Comput. Graph. Image Process, 1989. – Vol. 46. – p. 82–95.
14. Volotovskii, S.G. Machine Vision System for Registration of Oil Tank Wagons / Volotovskii S.G., Kazanskii N.L., Popov S.B., Khmelev R.V. // Pattern Recognition and Image Analysis, 2005. – Vol. 15(2). – pp. 461-463.
15. RF Patent 2336563, MPK G 06 K 9/00, 9/62, 9/66. A technique for recognizing the fluid jet disruption in images / N. L. Kazanskiy, N. Ye. Kozin, S. B. Popov, V. A. Fursov; Patentee: IPSI RAS, # 2006115915/09, appl. filed 05.10.06; published. 10.20.08, Bull. # 29.

---

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20