УДК: 004.932, 004.855

Using multiple video-information representations in automatic image-analysis systems

Publication in Journal of Optical Technology

For Russian citation (Opticheskii Zhurnal):

Потапов А.С. Использование множественных представлений видеоинформации в системах автоматического анализа изображений // Оптический журнал. 2012. Т. 79. № 11. С. 16–20.

Vasil’ev V. N., Gurov I. P., Potapov A. S. Modern video informatics: problems and prospects [in English] // Opticheskii Zhurnal. 2012. V. 79. № 11. P. 16–20.

For citation (Journal of Optical Technology):

A. S. Potapov, "Using multiple video-information representations in automatic image-analysis systems", Journal of Optical Technology. 79(11), 689-692 (2012)

Abstract:

This paper analyzes the necessity of using many information representations simultaneously in image-processing and -analysis systems. The difference between Kolmogorov-complexity and algorithmic-probability criteria when solving induction problems and making decisions is investigated. It is shown that making the optimum decisions (for example, in recognition or prediction problems) requires the use of many representations of information, in terms of which alternative descriptions of the images are constructed. A representational algorithmic-probability criterion is derived for determining the optimum set of representations from a given selection of images.

Keywords:

image analysis, information presentation, information theory, algorithmic complexity.

OCIS codes: 150.1135

References:

1. D. Marr, Vision: A Computational Investigation into the Human Representation and Processing of Visual Information (W. H. Freeman, San Francisco, 1982; Radio i Svyaz’, Moscow, 1987).
2. A. S. Potapov, “Choosing image representations by minimizing their representational description length,” Izv. Vyssh. Uchebn. Zaved. Prib. 51, No. 7, 3 (2008).
3. V. A. Kovalevski˘ı, “Local versus global decisions in image recognition,” Proc. IEEE 67, 745 (1979).
4. R. J. Solomonoff, Does Algorithmic Probability Solve the Problem of Induction? (Oxbridge Research, Cambridge, Mass., 1997).
5. R. J. Solomonoff, “Algorithmic probability, heuristic programming and AGI,” in Proceedings of the 3rd Conference on Artificial General Intelligence (AGI-2010), Lugano, Switzerland, March 5–8, 2010, pp. 151–157.
6. J. Poland and M. Hutter, “MDL convergence speed for Bernoulli sequences,” Stat. Comp. 16, 161 (2006).
7. E. Bauer and R. Kohavi, “An empirical comparison of voting classification algorithms: bagging, boosting, and variants,” Mach. Learn. 36, 105 (1999).
8. A. S. Potapov, “Synthetic pattern-recognition methods based on the representational minimum description length principle,” in Proceedings OSAV’2008, The 2nd International Topical Meeting on Optical Sensing and Artificial Vision, St. Petersburg, Russia, 12–15 May, 2008, pp. 354–362.

9. R. E. Schapire, Y. Freund, P. Bartlett, and W. S. Lee, “Boosting the margin: a new explanation for the effectiveness of voting methods,” Ann. Stat. 26, 1651 (1998).