Home – Home – Jornals – Avtometriya 2018 number 5

A combinatorial-discrete approach to solving continuous probabilistic problems arising in the analysis of random point structures is proposed. It is based on replacement of computational schemes based on calculations of cumbersome multidimensional integral expressions by schemes that do not require the use of the apparatus of differential and integral calculus. Another distinctive feature of the study is the systematic use of the generalized Catalan numbers from the multidimensional extension of the classical Catalan sequence for ranking random mutually dependent sequences.

The problem of detecting small-size objects in noisy images in the absence of information on the probability distributions of the magnitudes of the signals observed at points in the area of the object and at points of the complex (random) background surrounding the object is considered. Appropriate critical values are obtained for the relatively recently introduced new non-parametric statistical test in the case of detection of a small-size object. The powers of the new test and the well-known two-sample Wilcoxon test are compared for the case of exponential distributions.

Correction algorithm for digital elevation models derived from remote sensing of the Earth's surface are analyzed. The accuracy of the ASTER GDEM2, SRTM X -band, and ALOS DMS global digital elevation models were analyzed, showing that ALOS DMS pictures have the smallest absolute and relative errors for different relief conditions (flat, hilly and heavily dissected plains) of the Novosibirsk region. A comparative analysis of the algorithms proposed by Wang and Liu, Plantchon and Darboux, Pelletier, and Tarboton was performed to eliminate artifacts on original satellite images associated with local topographic lows (depressions, pits). The smallest errors for different terrain conditions were obtained using the algorithm of Wang and Liu.

Modifications of a nonparametric pattern recognition algorithm corresponding to the maximum likelihood criterion with additional decision functions are considered. The synthesis of the proposed algorithms is based on the analysis of the ratios of the estimates of the probability density distributions of random variables in classes and their functionals with input thresholds. The choice of the thresholds is determined by specific features of the classification problem. The results obtained are applied for assessing the states of forest tracts on the basis of remote sensing data.

A filtration algorithm is proposed for detecting objects in images obtained by means of microscanning. An algorithm for calculating the filter for an object with a known shape is presented. The influence of changes in the object shape induced by a subpixel displacement of the object with respect to the center of the photosensitive element on the filtration results is studied. An approach to choosing the filter shape with allowance for object shape changes is tested. Results of numerical simulations are reported.

Problems of determining the spatial orientation parameters of solids have been considered. Known results of mathematical modeling of a three-component flux-gate azimuth sensor have been analyzed. Mathematical models of a similar sensor azimuth with arbitrarily oriented fluxgates have been developed and analyzed.

The paper presents a brief analytical review of papers dealing with the use of the simplest binary-phase diffraction structures in multifocal artificial eye lenses (intraocular lenses) to increase the focal area. The results of theoretical studies of the influence of spherical aberrations of the eye, the optical power of the diaphragm (diameter of the pupil), and the porosity of the structure on the depth of focus (limits of clear vision) are given. The distribution of intensity in diffraction orders (foci) in the simplest binary diffractive-refractive intraocular lenses is analyzed.

Methods are proposed to improve the reliability of interference measurements of surface nanotopography with sharp height gradients that lead to ambiguity in determining the phase of interference signals. The effect of the total error of measurement on the range of multiwave measurements is considered. The results of field experiments demonstrating an increase in the range of measurements of nanotopography by the proposed methods in comparison with single-wave measurements are given.

This paper describes the autocollimation optical scheme of a fluorimeter that makes it possible to detect simultaneous reactions in all tubes of a microplate. It is shown that image focusing is carried out by a spherical mirror that excludes background glare and chromatic aberration. The operation of the fluorimeter is demonstrated by detecting a signal from the microplate with 96 and 384 tubes.

A new class of capacitive engines based on nanometer gaps between the electrodes and possessing specific powers much higher than that of inductive engines is described. Specific features of their operation are analyzed. The specific energy generated in a cycle of electromechanical conversion is estimated. The forces and powers of the considered engines are found to exceed the corresponding parameters of inductive engines by 2-3 orders of magnitude, which offers a possibility of using them in numerous applications.

This paper demonstrates the applicability of atomic-power microscopy for studying the elastic properties of suspended semiconductor structures on the basis of relatively thick GaAs/AlGaAs membranes in the case where the stiffness under study significantly exceeds the cantilever stiffness of an atomic-power microscope, which is confirmed by the agreement between the experimentally determined values of both relative and absolute stiffness measured at different structural points and theoretical values.

Results of an experimental study of a thin-film pyroelectric detector of radiation based on barium-strontium niobate and used for detecting nanosecond radiation pulses are reported. The possibility of recording radiation pulses with duration of 30 ns with the sensitivity of 1 V/W (NEP = 8 · 10^-7 W/Hz^1/2) is demonstrated.

A modified method of fuzzy clustering is proposed for determining the aerodynamic characteristics of an aircraft from flight test data. This approach allows one to describe the aerodynamic characteristics of an aircraft in the form of a black box model with inputs in the form of telemetry data, such as accelerations, angular velocities, thrust, and dynamic pressure, and with outputs in the form of dimensionless aerodynamic coefficients of forces and moments. Results of modeling in the MATLAB/Simulink environment are reported.

A method for automatic determination of combustion regimes using flame images on the basis of the tagged data of a trained convolutional neural network is under consideration. It is shown that the accuracy of regime classification reaches 98 % on the flame images of a gas burner. The results of the operation of the convolutional neural network and classification using different linear models are compared.

This paper describes the problem of controlling a quadcopter movement in cargo transportation on suspension. The suspension is a weightless rod pivotally fixed at the quadcopter mass center with a cargo on its end. Equations of motion for a two-dimensional model are obtained using Lagrange equations of the second kind, which serve as a basis for a model revealing the suspension oscillations that destabilize the system. A controller based on the structural synthesis method and including the deflection angle of the rod from the vertical is proposed. The efficiency of the algorithm is confirmed by the numerical simulation results.

This paper describes the synthesis of an adaptive system for automatic control of the quick drier of an organic waste disposal device. The application of an adaptive controller for loading the dryer with wet fuel.