V.I. Terekhov - Ph.D. (Eng.), Associate Professor, Department of Information Processing and Control Systems, Bauman Moscow State Technical University E-mail: terekchow@bmstu.ru I.M. Chernenkiy - Post-graduate Student, Department of Information Processing and Control Systems, Bauman Moscow State Technical University E-mail: cheivan@mail.ru

This paper discusses anamorphosis method application issues for problems of one or more spatial indicators. Taking them into account, a decision maker develops an option for the informed decision. The reason of this approach is that the anamorphosis method activates decision maker-s visual-imaginative arrangements in a complex situation and allows visua-lizing spatio-temporal indicators of several processes or phenomena described by the indicators. Herewith, one or several indicators are taken uniformly distributed over the whole space o decision making and all other indicators, which are related to them, are analyzed on their background. The essence of the method is that original visual images constructed in the Euclidean metric based on a selected indicator are converted (anamorphosed) into two-dimensional visual images. Their internal structure changes in a special way when the distribution of the selected indicator becomes uniform, while maintaining topological similarity with the original visual image. The paper presents the mathematical formulation of the problem and features of the anamorphosis algorithm. As a basic option of the iterative algorithm, an algorithm for constructing an anamorphose, which is efficiently working in real time, is shown. Based on the algorithm, two options of the modified anomorphosis algorithm were developed and studied in detail as well as their performance with an integral anamorphic indicator. The indicator is a convolution of several indicators of different physical nature, which take into account their importance and the balance in the Euclidean metric. The first option of the modified algorithm works when there is time for decision making with any desired accuracy. It uses sigmoidal functions to normalize the parameters to prevent topology violations of the original visual image. The second option allows significantly decrease time for the anamorphose obtaining at given accuracy reduction. For these cases, an artificial neural network with one hidden layer for calculations was used. The network was trained on examples retrieved from the first option usage results of the modifies anomorphosis algorithm. The obvious advantages of the anamorphosis method for developing an informed decision are: - decreasing the problem dimension to the number of local indicators different physical nature, folded into the integral anamorphic indicator; - possibility for the visual modeling and revealing hidden behavior patterns of various indicators of the problem, im-plicitly depend on the local or integral anamorphic indicator; - possibility to build scenarios based on the spatio-temporal anamorphose analysis taking into account processes re-lated to a rapid the local or integral anamorphic indicator evolution; The conclusion of the paper shows that the anamorphosis is a promising method, allowing developing informed de-cisions in various operation modes of the decision maker.

 

1. Pospelov D.A. Fantazija ili Nauka. Na puti k iskusstvennomu intellektu. M.: Nauka. 1982.
2. Glezer V.D., Cukerman I.I. Informacija i zrenie. M.: AN SSSR. 1961. 47 s.
3. Zenkin A.A. Kognitivnaja kompjuternaja grafika / Pod red. D.A. Pospelova. M.: Nauka. 1991. 192 s.
4. Ursul A.D. Informacija. Metodologicheskie aspekty. M.: Nauka. 1971. 130 s.
5. Gusejjn-Zade S.M., Tikunov V.S. Anamorfozy: chto ehto takoe - M.: EHditorial URSS. 1999.
6. Bezdek J.C. What is computational intelligence - // In Computational Intelligence Imitating Life. Zurada, J. M., R. J. M. II, and C. J. Robinson, Eds. // IEEE Press. New York. 1994. R. 1-12.
7. Zajjchenko JU., Zgurovskijj M. Sistemnyjj analiz napravlenijj vychislitelnogo intellekta. International Journal «Information Content and Processing». 2014. V 1. № 3. P. 220.
8. Tobler W.R. A transformational view of cartography // Amer. Cartogr. 1979. V. 6. № 2. P. 101-106.
9. Kadmon N., Shlomi E. A polifocal projection for statistical surfaces // The Cartographic Journal. 1978. V. 15. № 1 P. 36-41.
10. Petrov P.V., Serbenjuk S.N., Tikunov V.S. Analiticheskijj sposob sozdanija anamorfirovannykh kartograficheskikh izobrazhenijj // Vestnik Moskovskogo un-ta. Ser. Geografija. 1983. № 2. S. 56-63.
11. Selvin S., Merrill D., Sacks S., Wong L., Bedell L., Schulman J. Transformations of maps to investigate clusters of disease. Lawrence Berkeley Laboratory, Univ. of California. 1984. 33 p.
12. Dougenik J.A., Chrisman N.R., Niemeyer D.R. An algorithm to construct continuous area cartograms // Prof. Georg. 1985. V. 37. № 1. P. 75-81.
13. Terekhov V.I. Primenenie kognitivnojj kompjuternojj grafiki v sistemakh podderzhki prinjatija reshenija dolzhnostnykh lic organov voennogo upravlenija. Monografija. M.: Izdatelstvo «Obshhevojjskovaja akademija VS RF»/ 2012. 150 s.
14. https://en.wikipedia.org/wiki/Cartogram, (Data obrashhenija: 10.10.2016).