Математика, механика, информатика. Рубрика в журнале - Сибирский аэрокосмический журнал
A comprehensive evolutionary approach for neural network ensembles automatic design
Статья научная
A new comprehensive approach for neural network ensembles design is proposed. It consists of a method of neural networks automatic design and a method of automatic formation of an ensemble solution on the basis of separate neural networks solutions. It is demonstrated that the proposed approach is not less effective than a number of other approaches for neural network ensembles design.
Бесплатно
A mathematical model of oil price assessment
Статья научная
The article deals with the development of the mathematical model of oil price assessment. The methodological foundation is determined for developing the mathematical model: two axioms stating the unique properties of oil as a commodity. The first one claims that oil is a commodity being determined at auctions and is not related to its value as a measure of abstract labor invested, the second axiom states that the markdown in oil price will not cause the increase in the demand for it, as the demand is determined only by the economy state of the demander. Among the factors of oil pricing an imbalance of oil supply and demand in the world market is chosen to be the dominant factor. The mathematical model is represented in two models. The first one assumes that for any excess of supply over demand, the price of oil tends to zero, i. e. for a sufficiently large number of auctions it becomes lower than any predefined level. The second theorem states that in the case of the excess of demand over supply oil price tends to infinity (a finite number of sessions exceeds any predefined level) in case of the dominance of imbalance. The most likely forecast resulting from the hypothesis that the developed mathematical model is correct is the trend of the price decrease reaching its extremely low level and a further transition into a long-term period characterized by the price increase trend.
Бесплатно
A method of image segmentation with the help of areas growing and multiscale analysis
Статья научная
In this article we analyzed advantages and disadvantages of existing methods of image segmentation. The development of an original algorithm of segmentation which uses the method of areas growing and multi scale analysis is presented. The capabilities of this method in different images segmentation are researched.
Бесплатно
A multidimensional analog of the Cooley-Tukey FFT algorithm
Статья научная
In this article a recurring sequence of orthogonal basis in the n-dimensional case has been applied to derive formulas of n-dimensional fast Fourier transform algorithm, which uses Complex multiplication and nN n log 2 N complex addition; where N = 2 s – is a number of counts on one of the axes.
Бесплатно
A new fast face detection technique
Статья научная
The problem of human face detection in a natural or artificial environment has always been among the highest priorities for researchers working in the field of computer vision systems and artificial intelligence. An effective face detection system should provide high percentage of correct detections, and low false detection rate in short time. Viola-Jones method is one of the best algorithms in terms of speed/quality ratio. However, this method in many cases gives a large number of false detections. The color of human skin is one of the features that helps to make face detection. The presence of the color information improves the efficiency of face allocation; narrows the search area and reduces the number of false detections and processing time of the input images. This paper solved is the problem face area detection, based on two new methods. The first technique uses the image pixel skipping process instead of testing each pixel to label it as skin or non-skin by using RGB color space. Second technique uses YCbCr color space and block approach which divides the image into blocks, the size of each block is 3×3 pixels, then we check the central pixel. If the central pixel satisfies the skin criteria, the whole block will be considered as a skin. Finally, we applied Viola-Jones algorithm to detect faces. The experimental results presented in the paper shows that the proposed algorithms provide high speed detection at low false error rate.
Бесплатно
Статья научная
In order to simulate the Vaganov–Shashkin model for seasonal growth and tree-ring formation, a solution algorithm for the parameterization problem of the model is being proposed in cases, when a modulation is possible. The algorithm is realized as dll-library (or as a text file), tested on extensive data. A concept of difference in criterion between the actual tree-ring chronology and its model is introduced. Two new difference criteria are developed.
Бесплатно
A practical approach to software portability
Статья научная
This paper describes an approach to porting onboard software for communication and navigation satellites to new platforms that use various onboard computers and devices. The approach relies on the target(onboard) and the tool software stratification and strong typing and the Modula-2 programming language especial features.
Бесплатно
Статья научная
In this article we have considered a possibility to use the commercial XBO xenon lamps to create a source of radiation, integrated in solar simulators. We have conducted experimental studies of photonic lamp performance.
Бесплатно
About multiagent system applications for speech recognition problem
Статья научная
In this paper we suggest two different multi agent systems for speech recognition problem. The multi agent systems (MAS) are becoming very popular because of their flexibility and applicability to complex problems. The system is based on functioning of different agents that forms the system and interacts with each other. The main profit of using multi agent approach is that every agent can be described as a simple subsystem and the whole initial task can be solved with automatic and autonomous agent actions, interactions and decision making. So the main problem can be reduced to behavior rule base tuning.
Бесплатно
About non-parametric identification of T-processes
Статья научная
This paper is devoted to the construction of a new class of models under incomplete information. We are talking about multidimensional inertia-free objects for the case when the components of the output vector are stochastically dependent, and the character of this dependence is unknown a priori. The study of a multidimensional object inevitably leads to a system of implicit dependencies of the output variables of the object from the input variables, but in this case this dependence extends to some components of the output vector. The key issue in this situation is the definition of the nature of this dependence for which the presence of a priori information is necessary to some extent. Taking into account that the main purpose of the model of such objects is the prediction of output variables with known input, it is necessary to solve a system of nonlinear implicit equations whose form is unknown at the initial stage of the identifica- tion problem, but only that one or another output component depends on other variables which determine the state of the object. Thus, a rather nontrivial situation arises for the solution of a system of implicit nonlinear equations under condi- tions when there are no usual equations. Consequently, the model of the object (and this is a main identification task) cannot be constructed in the same way as is accepted in the existing theory of identification as a result of a lack of a priori information. If it was possible to parametrize the system of nonlinear equations, then at a known input it would be necessary to solve this system, since in this case it is known, once the parameterization step is overcome. The main content of this article is the solution of the identification problem, in the presence of T-processes, and while the pa- rametrization stage can not be overcome without additional a priori information about the process under investigation. In this connection, the scheme for solving a system of non-linear equations (which are unknown) can be represented in the form of some successive algorithmic chain. First, a vector of discrepancies is formed on the basis of the available training sample including observations of all components of the input and output variables. And after that, the evalua- tion of the output of the object with known values of the input variables is based on the Nadaraya-Watson estimates. Thus, for given values of the input variables of the T-process, we can carry out a procedure of estimating the forecast of the output variables. Numerous computational experiments on the study of the proposed T-models have shown their rather high effi- ciency. The article presents the results of computational experiments illustrating the effectiveness of the proposed tech- nology of forecasting the values of output variables on the known input.
Бесплатно
About parametric identification algorithms of discrete-continuous processes
Статья научная
Researches presented in the paper are devoted to parametric modelling of multidimensional processes of discrete- continuous type in the condition of priori information lack. Similar processes occur in the space industry, for example, in the manufacture of products based on electronic components. The article considers multidimensional processes with unknown mathematical description. Using parametric approach, we choose the structure of investigated process with the accuracy to within parameters, and the next step is to estimate the model parameters from the available sample of observations of the process input and output variables. The paper examines the case when due to the lack of priori knowledge about the object an error is allowed at the stage of parametric structure choosing. The relative approxima- tion error is used to estimate the model accuracy, which shows the difference between model and object outputs. A comparative analysis of several parametric models for one investigated object is carried out is. Using the method of least squares we obtain estimates of the parameters. The paper presents the results of a series of computational experiments illustrating the dependence of the modelling error on the object noise level, as well as on the sample size of observations of the input and output variables. One of the obvious parametric models advantages is the ease of its applying. However, if the dimension of the input variables vector is high, the process has a complex structure, and there is no priori information about the object structure, then it is difficult to use parametric methods. In this case, it is advisable to use nonparametric identification methods. In this paper we use a nonparametric estimation of the regression function on observations of Nadaraya-Watson as an estimate of the process output variable. However, such estimates require a large number of initial data, also they are sensitive to various kinds of defects in the initial samples of observations. Besides that, the paper compares nonpara- metric model with parametric one for the investigated process.
Бесплатно
About the analysis of the pulse-width system with feedback
Статья научная
In the article research results of the pulse-width system (PWS) captured by a negative feedback circuit have been depicted. Based on an asymptotic method of order decrease in the linear part of system, we have offered a technique for decreasing the PWS to an equivalent nonlinear pulse-amplitude system for which known methods of research are applied.
Бесплатно
Статья научная
Using feature selection procedures based on filters is useful on the pre-processing stage for solving the task of data analysis in different domains including an air-space industry. However, it is a complicated problem, due to the absence of class labels that would guide the search for relevant information. The feature selection using “wrapper” approach requires a learning algorithm (function) to evaluate the candidate feature subsets. However, they are usually performed separately from each other. In this paper, we propose two-stage methods which can be performed in supervised and unsupervised forms simultaneously based on a developed scheme using three criteria for estimation (“filter”) and multi-criteria genetic programming using self-adjusting neural network classifiers and their ensembles (“wrapper”). The proposed approach was compared with different methods for feature selection on tree audio corpora in German, English and Russian languages for the speaker emotion recognition. The obtained results showed that the developed technique for feature selection provides to increase accuracy for emotion recognition.
Бесплатно
About the model of the active system
Статья научная
The problem of the active system modeling is researched. There are three sources of information to construct the models. The first source is the theory of the problem, the second one is experts in the field ofproblem, and the third one is data bases. The method considered in this article uses the third way. The nonparametric algorithm for modeling active element in the system is proposed. The results of the computer experiments are published.
Бесплатно
About torsion of parallelepiped around three axis
Статья научная
The theory of limit state deals with statically determinate condition of solids. In this case the system is closed due to extreme conditions, such properties of matter such as viscosity, elasticity, etc. cannot influence the limit state. In other words, when reaching the limit state the nature of the relationship between stress and strain has no effect on the ulti- mate state. The study of such systems has been consistently pursued by D. D. Ivlev and his coauthors. To the equilib- rium equations they attached two or an equation relating the components of the stress tensor. This led to the closure of the system of equilibrium equations. In the theory of plasticity equations, which are closed with a single yield stress are studied well. The most well-known system describing the ultimate state of deformable bodies are well-studied equations describing the torsion of the plastic bodies, the two-dimensional stationary problem of the theory of plasticity. The arti- cle discusses some other systems of equations which are closed only by one equation of flow, which corresponds to the classical theory of plasticity. It is assumed that the components of the velocity vector depend only on two spatial coor- dinates. In addition, for the component of velocity vector conditions of deformations compatibility are performed identi- cally. The constructed systems can be used to describe the twisting of the parallelepiped around the three orthogonal axes. For the constructed system of equations point group symmetries, conservation laws have been found. It is shown that the system allows 8 -dimensional Lie algebra. On the basis of the symmetry group some classes of invariant solutions of rank 1 have been constructed. They depend on arbitrary functions of one variable. It is shown that these solutions can be used to describe plastic torsion of a parallelepiped around three orthogonal axes. It is shown that the system admits infinite series of conservation laws. The concluding paragraph describes the construction of elastic solutions to the problem. It is shown that it boils down to finding three harmonic functions.
Бесплатно
Active metamaterial on the base of integral NEMS-structures
Статья научная
The conception of integrated nanoelectromechanical systems (NEMS) formation method is considered. The method is based on original combination of self-organizing and self-aligning processes. The functionality of proposed NEMS-structures and possible applications of nanomaterial which constituted by two-dimensional array of such structures are discussed. The results of experiments directed to proposed NEMS-technology realization are led.
Бесплатно
Статья научная
A program-realized coordinated algorithm of choice for automatic excitation regulator settings has been developed. This algorithm is based on the resultant theory and applies a mathematical model which is synthesized by experimental frequency characteristics of an electric power system.
Бесплатно
Algorithms for calculating complex indicators in dynamic structures of data representation
Статья научная
This paper presents algorithms for calculating complex indicators on set factual data, represented in dynamic structures with the application of the graph theory.
Бесплатно
Статья научная
A new algorithm for the solution of complex constrained optimization problems based on the probabilistic genetic algorithm with optimal solution prediction is proposed. The efficiency investigation results in comparison with standard genetic algorithm are presented.
Бесплатно
An algorithm for an object grasping by a manipulator in an unknown static environment
Статья научная
An algorithm for a n-link manipulating robot (MR) control in an environment with unknown static obstacles is considered. A theorem is proved which states that following the algorithm a MR in a finite number of steps will either grasp an object or will give a proved conclusion that an object cannot be grasped in any configuration.
Бесплатно
