Home – Home – Jornals – Avtometriya 2003 number 6

A linear model and a nonlinear model of perceptional space created by human mind as a result of visual perception of the surrounding world are proposed. The linear model corresponds to the monocular vision and satisfies the laws of projective geometry and Emmert

Circular scanning laser writing systems are effectively used in fabrication of diffractive optical elements, micro-optics, coded disks, and limbs. However, the accuracy of microstructure fabrication by means of such systems depends substantially on the radial spindle runout that causes deviation of the writing beam trajectory from a circular form. Knowing the dependence of this deviation on the angle of rotation, one will compensate for the influence of spindle runout by correcting the radial beam position. A new photoelectric method for measuring the writing beam trajectory is proposed for circular scanning systems. Results of its application to correct the beam position by an electromagnetic deflector are discussed.

The total and minimized instrumental errors of a diffraction common-path interferometer are analytically investigated. The interferometer is based on a zone plate and intended to measure the shape of surfaces and mirrors of the second order. In the scheme under analysis, the interfering wave fronts are noninverted wave fronts. It is shown that the character of instrumental error variation relative to the coordinates of the measured surface does not depend on the value of its conic parameter. When the focal length of a zone plate is much smaller than the surface curvature radius, spherical articles are controlled by the null method whereas in the case of aspherical articles, preliminary calculated corrections are introduced to the measurement results. The obtained relationship between the value of mirror displacement and the instrumental error allows one to calculate the admissible accuracy of mirror position.

Distortions of geometrical sizes of the simplest diffractive structures written by the thermo-chemical method in a chromium film (80

A method for refining the parameters of the absolute satellite camera orientation from a series of space pictures with the common target point is proposed and investigated. The method is based on simultaneous processing of three-dimensional coordinates of control (polygon) objects and projective coordinates of their images on a sequence of frames. An algorithm of singular value decomposition of the data matrix is used for precise estimating the angles of orientation of a satellite camera. The error of estimating the absolute orientation angles is investigated as a function of the number of surface reference objects and the noise level in the projective coordinates of images of the objects. The precision refinement of camera orientation allows sufficiently accurate and adequate relief reconstruction from a sequence of conjugate points on surface pictures. Simulated characteristics of the errors of estimating the coordinates of surface elements are presented.

An updated version of the sampling theorem is considered under periodically nonuniform and uniform sampling of a bandwidth-unlimited signal. The obtained relationships for signal reconstruction will reduce the variance of its reconstruction error.

A control and measurement information system for investigating the spectral and current-voltage characteristics of semiconductor devices is created. The influence of nonequilibrium surface charge accumulation on the spectral characteristics of silicon bifacial solar cells with a shallow nonsymmetrical n

To form the tomographic patterns of underwater objects obtained by the two-sided translucence method (bistatic sonar) or numerical modeling, it is proposed to apply convolution methods or a phase change method. It has been shown in practice, that these methods are most suitable not in object image construction but in directed summation transformation) of data obtained on the receive system recorder and processed in the analog-to-digital converter of the computer system. It is shown that the immediate conversion from the space of projections to the image plane is implemented by the inverse Radon transform or by reconstruction from filtered projections. It is noted that as the emitted signal frequency decreases, these methods give negative results. Hence, in this case it is recommended to apply the phase change method that allows implementation of frequency-phase processing of the detected signals.

The possibility of using various types of intrinsic radiation of high-voltage electrical power equipment as diagnostic signals in estimating and predicting the technical state of high-voltage equipment (HVE) under operating voltage is studied. The problem of individual prediction of the state of electrical power equipment is considered. Methods and algorithms for solving this problem under conditions of the lack of information on statistical properties of possible states are examined. Requirements to the multi-channel system for acquisition, input, and computer processing of the signals of sensors of intrinsic HVE radiation are formulated.

The problem of optimal estimating the power spectrum of stochastic process in the base of discrete exponential functions is considered. An optimal weight function is synthesized. The function minimizes the mean square error of estimation under a given limitation on the bias error. An adaptive procedure of modifying the power spectrum estimator under mutual cancellation of bias error components is developed. The procedure will increase the accuracy of spectral estimation.

Automation of decoding monochromatic images regardless of their content requires obligatory solution of some tasks. The main tasks are to describe a real scene, determine the volume of information on the scene, which can be obtained from its image, select the features, and perform segmentation. The present paper, within the scope of the theoretical-probabilistic approach, is an attempt to analyze mathematical and methodical problems arising in solving the above-listed tasks.

The problem of collision detection for functionally defined objects is considered. The algorithm for collision detection for functionally defined objects of different complexity during a constant time is discussed.