Метод направленного перебора альтернатив в задаче распознавания полутоновых изображений
The article discusses development of the segmented characters classifier of the Russian alphabet a nd of the Arabic numerals on the basis of block neural network structures including the plurality of blocks for each individual character recognition and for the synthesis block decision. Keywords: pattern recognition, neural network, training of neural n etworks, base of hand - written characters, recognition of hand - written characters
The paper is devoted to the description of two approaches for recognizing railway roadbed profiles, obtained with the help of laser scanning. The first approach is based on the identification of similar parts of comparing profiles, presented by their polygonal representations. The second approach uses weighted functional metrics, where it is possible to process incomplete data and then to make a choice on the set of preferences.
The problem of the choice of algorithms parameters in automatic image recognition is put and solved by ensemble classifiers construction using the maximum posterior probability principle. The new criterion of parameters choice is strictly synthesized for Kullback-Leibler information discrimination and modern SIFT (Scale-Invariant Feature Transform) method of object recognition. The program and results of experimental research in a problem of face recognition with widely used databases (Yale, AT&T) are presented. It is shown that the proposed criterion allows to achieve recognition accuracy equal to the algorithm with the best parameters set, and not only for Kullback-Leibler information discrimination, but also for other popular distances (Euclidean metric, Kullback information divergence).
Studied is a possibility of increasing the accuracy of diagnostics by examining a number of diagnostic rules as a set of expert assessments, which allows one to combine them («mix of expert opinions»). Proposed is to use of the principle of minimum-information-mismatch in Kullback - Leibler metric to highlight the rule most appropriate for classification of a particular object. Program and results of experimental study are presented in the problem of automatic recognition of gray-scale images. It is shown that the developed approach can significantly improve the quality of diagnostics.
A novel method for evaluating classification reliability is proposed based on the discernibility of a pattern’s class against other classes from the pattern’s location. Use of three measures of discernibility is experimentally compared with conventional techniques based on the classification scores for class labels. The classification accuracy can be drastically enhanced through discernibility measures by using the most reliable – “elite” – patterns. It can be further boosted by forming an amalgamation of the elites of different classifiers. Improved performance is achieved at the price of rejecting many patterns. There are situations where this price is worth paying – when the non-reliable accuracy rates lead to the need in manually testing of very complex technical devices or in diagnostics of human diseases. Contrary to conventional techniques for estimating reliability, the proposed measures are applicable on small datasets as well as on datasets with complex class structures where conventional classifiers show low accuracy rates.
This book constitutes the refereed proceedings of the 9th International Conference on Cellular Automata for Research and Industry, ACRI 2010, held in Ascoli Piceno, Italy, in September 2010. The first part of the volume contains 39 revised papers that were carefully reviewed and selected from the main conference; they are organized according to six main topics: theoretical results on cellular automata, modeling and simulation with cellular automata, CA dynamics, control and synchronization, codes and cryptography with cellular automata, cellular automata and networks, as well as CA-based hardware. The second part of the volume comprises 35 revised papers dedicated to contributions presented during ACRI 2010 workshops on theoretical advances, specifically asynchronous cellular automata, and challenging application contexts for cellular automata: crowds and CA, traffic and CA, and the international workshop of natural computing.
The manual is intended for students of Department of computer engineering MIEM HSE. In the textbook based on the courses "Economics of firm" and "the development strategy of the organization." Discusses the key conceptual and methodological issues of the theory and practice of Economics and development planning of the organization. The use of textbooks will enable students: to analyze key performance indicators, and use the tools of strategic analysis with reference to concrete situations in contemporary Russian and international business. Special attention is paid to the methods and systems of information support of the life support functions of business organizations and management methodology of innovation and investment. An Appendix contains source data for analysis of competition in a particular industry.
The paper provides a number of proposed draft operational guidelines for technology measurement and includes a number of tentative technology definitions to be used for statistical purposes, principles for identification and classification of potentially growing technology areas, suggestions on the survey strategies and indicators. These are the key components of an internationally harmonized framework for collecting and interpreting technology data that would need to be further developed through a broader consultation process. A summary of definitions of technology already available in OECD manuals and the stocktaking results are provided in the Annex section.
Over the last two decades national policy makers drew special attention to the implementation of policy tools which foster international cooperation in the fields of science, technology, and innovation. In this paper, we look at cases of Russian-German collaboration to examine the initiatives of the Russian government aimed at stimulating the innovation activity of domestic corporations and small and medium enterprises. The data derived from the interviews with companies’ leaders show positive effects of bilateral innovative projects on the overall business performance alongside with major barriers hindering international cooperation. To overcome these barriers we provide specific suggestions relevant to the recently developed Russian Innovation Strategy 2020.