Home – Home – Jornals – Avtometriya 2003 number 4

A new approach to image isoline approximation by smoothing splines is proposed. The approach eliminates the stage of initial isoline parametrization and allows one to prescribe

A method is developed and described for fast solving in the spatial domain the problem concerned with initial signal (image) reconstruction by a set of low-resolution images that differ by the mutual coordinate shift. The proposed algorithms and the computational scheme implemented on their basis are constructed in such a way that a signal with the minimal energy (dispersion) is selected from the whole class of digital signals satisfying the set of observed data.

Dependences of the efficiency while recognizing noisy static and moving test objects on the duration of their presentation are experimentally studied. Investigation is conducted for different observation conditions. It is found that with increasing duration of observing the dynamic images of test objects in the presence of a white noise the efficiency decreases at first. It is underlined that upon achieving the minimal value that takes approximately 120 ms it grows and then upon achieving the maximal value taking approximately 400 ms it decreases monotonically. Mathematical modeling of the processes of accumulation of the image signal component in the visual memory is fulfilled.

A method for evaluation of the frequency of impact impulse appearance and qualitative analysis of vibrosignal components is described. It is based on nonstandard wavelet transform. The method is compared with the Fourier transform and its capabilities, advantages, and drawbacks are evaluated. The method is used to analyze a real signal from a test machine with a defective rolling bearing. Analysis of the real signal leads to the conclusion that it is insufficient to consider real signals by analyzing only the Fourier spectrum.

An operative method of abnormal measurement rejection for multidimensional dynamic systems is developed. The method is robust to systematic measurement errors. It does not require information on the value and sign of systematic errors, and also on statistical characteristics of abnormal measurements, and can be used in various technical fields for measurement information processing.

The possibility of applying the sliding mode method to synthesis of an automatic extremum seeking control system for dynamic plants with a static extreme characteristic is discussed. It is recommended to transform preliminary the control plant model by adding a new variable. For the variable it is proposed to use an estimate of the output characteristic gradient. It is shown that this transformation makes it possible to reduce the initial problem of synthesizing the extremum seeking system to a stabilization problem usual for the automatic control theory. The main properties of a closed-loop system are considered and some advises for control algorithm implementation are presented.

The problem of spatial localization of the coherent signal source energy is investigated within the scope of the quadratic and nonquadratic regularization theory on the basis of l₂-norms in the conditions of minimal a priori information on the form, frequency-time regions of broadband signal existence, and statistical characteristics of disturbances. A numerical experiment using an antenna array of a nonlinear spatial configuration proved the efficiency of the proposed algorithms for digital reconstruction of the angular energy distribution in the conditions of multipath propagation.

It is shown that for the optimal (by the minimum mean-root-square error) prediction of the moving average process of the q-order the predictability time can exceed the correlation time by a factor of in q the case of a certain combination of the moving average coefficients.

Estimation of intensity of local plastic deformation developing under conditions of fatigue fracture is suggested on the basis of a number of numerical characteristics obtained by analyzing a two-dimensional power spectrum of a series of optical images of the surface. It is shown that the presence of a fatigue crack as a region of low information on the image does not exert any significant effect on the calculated results of energy characteristics of the power spectrum. A comparison of the calculated results with the results of fractal dimension calculations shows qualitative similarity of these approaches, which is based on the multidimensionality of measurements implemented in both of the cases.

An expression is given for a normalized moment (determined in an arbitrary direction) of an energetic spectrum of a gradient image of an arbitrary order. It is shown that the matrix form of this expression makes it possible to obtain analytical expressions for the main moments of the spectrum and the main values for the mean square frequencies of the gradient image and the corresponding directions. It is demonstrated that the estimates of the characteristics can be obtained in the plane of space variables without performing the direct Fourier transform, which is important in solving the problems of image express analysis. Experimental results on a limited amount of test images are presented.

A cellular neural network (CNN) is known as a fine-grain parallel model capable to generate various types of autowaves. This property of CNN allows one to consider it as a new model of autowave processes in distributed active media and determines the necessity of further detailed investigation of the model. A CNN model modification that does not have the neuron bias parameter is described along with its properties. Techniques of using the external inputs for modifying the CNN properties and generating the autowave processes are determined.