Home – Home – Jornals – Avtometriya 2009 number 1

A method for constructing fast projection-variance algorithms based on convolution algorithms for calculating variances of Radon projections is proposed. It is shown that the use of bilinear interpolation of the convolution kernel and parabolic interpolation of quantized estimates of directional fields increases the accuracy of the algorithms by more than a factor of 2. Two methods for constructing directional fields on the basis of Radon projections are considered and compared. Examples are given showing the computational efficiency, accuracy, and noise resistance of the proposed algorithms.

An algorithm is proposed for detection, in a sequence of images, of small-scale objects moving with a known velocity vector, the initial positions of the objects being unknown. Random noise is suppressed by four-channel filtration based on estimating subpixel coordinates of the objects in the sequence of images. Results of numerical experiments are presented as a dependence of the detection probability on the false alarm level for different object sizes and input noise levels.

Methods are proposed to generate thredimensional surface images with improved angular resolution for low-altitude airborne multichannel radars based on the optimal reconstruction of reflection fields and narrow-band Doppler filtering.

Synthesis and analysis of the maximum likelihood algorithm of detecting an object extended in space are performed. The influence of the absence of a priori information on the velocity vector of object motion and image and background intensities on detection efficiency is studied.

A method based on invariants of in-plane motion of an object is developed for compensation of systematic errors of measurements in goniometric systems. Exact equations are derived for systematic errors of measurement of the object latitude with ignored fluctuating errors of  measurement. Analytical expressions are also derived to calculate the standard deviation of the error in estimating systematic errors being taken into account, which allow us to impose grounded requirements to tactical and technical characteristics of goniometric systems designed for observation of various objects.

The problem of automatic segmentation of a stochastic time series into homogeneous data segments one cluster long is posed and solved according to the general formulation of the disorder problem. A new algorithm with cluster normalization by the variance of the generating noise is developed using an autoregression model and the minimum information divergence criterion. It is shown that the main advantage of the proposed algorithm over existing analogs is high dynamic properties. The algorithm was tested by analyzing stock market dynamics in the United States and Russia. Estimates of the admissible (threshold) level of disorder in a time series within one cluster are obtained using the Kullback-Leibler information measure.

Nonparametric (kernel) estimation of a probability density function f(x) for a sample of finite size is considered using the -approach. The smoothness parameter β of the estimated probability density is introduced. For the case β > 2, it is shown that the convergence of the density estimate f_n(x) to the function f(x) can be improved by using alternating-sign weight functions (higher-order weight functions). Estimation of the derivatives of a function is briefly considered using the same approach.

The properties and areas of application of images formed by an array of light emitting diodes with modulated brightness observed in an oscillating mirror are considered. It is shown that these properties can be used to immerse large groups of spectators into virtual environment.

A new iterative algorithm of tomographic reconstruction of objects on the basis of projection data available in a limited range of angles only is proposed. The algorithm is based on calculating artificial projections in those directions where projection data are unavailable. By means of numerical simulations, it is verified that the algorithm developed ensured high quality of reconstruction up to the angular interval of 45-60

This paper describes the Spektran software system intended for automated analysis of object-attribute data tables which implements data mining algorithms based on a function of rival similarity (FRiS). The Spektran system is used to analyze a set of objects (microparticles of a substance) described by spectral characteristics. The following basic problems of data mining are solved: particle clustering by similarity of their spectra, selection of the subset of the most informative spectrum channels, identifying the classes to which particles and their mixes belong and some others.

A multichannel microlens integrated-optic coupler is designed. The coupler outputs the radiation through the substrate surface, thus, the signal detectors can be located on the surface. The procedures of coupler design, fabrication, and investigation are described.

Specific sources of the error of forming the angle measuring structure topology by raster scanning based on the thermochemical technology of latent image writing in chromium films are analyzed. Relationship between this type of error and the specific features of dynamics of chromium film heating by laser radiation is found. A structure writing correction algorithm is proposed and implemented. The algorithm ensures an accuracy of synthesizing angle measuring structures of about several seconds of arc.

A modified signal channel of holographic memory is proposed and studied theoretically and experimentally. Compared with a conventional optical scheme, the channel enables to decrease the spread in signal of reconstructed images of information elements, which is caused by their interference in the case of a restricted Fourier hologram aperture, increase the signal-background ratio, and reduce the sensitivity to misalignment of the image and the photodetector array.