Large-dimension image clustering by means of fragmentary synchronization in chaotic systems

The paper describes chaotic neural network improvements to enhance the quality of n-dimensional image clustering. New synchronization type—cluster fragmentary synchronization was found. To reveal fragmentary synchronization a new method is proposed. Benderskaya Elena Nikolaevna was born in 1969. She graduated from Automation and Computer Systems department of the St. Petersburg State Polytechnical University in 1993. In 1996 she received her candidate’s degree in the field of mathematical and computer modeling. Since 1996, she delivers lectures at the Faculty of computer science, SPbSPU. Her scientific interests include different aspects of artificial intelligence, namely, neural networks and adjacent fields (pattern recognition, fuzzy logic, evolution computations, and synergetics). She is the winner of the Informica 2006 All-Russian Grant Contest (FGU GNII ITT)—a contest between the leaders of scientific teams working in the field of telecommunication technology. Zhukova Sofya Vitalyevna was born in 1981. She graduated from Automation and Computer Systems department of St. Petersburg State Polytechnical University in 2004. In 2007, under E.N. Benderskaya’s supervision she defended her PhD thesis in the field of system analysis, control, and information processing (informatics). Since 2008 she holds the appointment of associate professor at the Graduate School of Management, St. Petersburg State University. Her scientific interests include control theory, internet technologies, self-organization theory, the theory of nonlinear systems, and the theory of neural networks. She was awarded a grant from the Government of the Russian Federation as a result of the All-Russia Contest of 2006–2007 for Postgraduates.