Home – Home – Jornals – Avtometriya 2016 number 1

The efficiency of a number of the classical methods of supervised classification of hyperspectral data is estimated by an example of discriminating the types of the underlying surface in natural and man-made areas. The minimum distance, support vector machine, Mahalanobis, and maximum likelihood methods are considered. Particular attention is paid to studying the dependence of the data classification accuracy on the number of spectral features and the way of choosing them in the above-mentioned methods. Experimental results obtained by processing real hyperspectral images of landscapes of various types are reported.

A model of image recording in the circular microscanning regime is considered. An algorithm that allows obtaining an image with a large number of samples from a set of smaller-size frames for a known path of the scanning array is proposed. Numerical simulations are performed to estimate the changes in the amplitude of a small-size object with a static position of the photodetector array and microscanning in the start/stop mode. The influence of the gap between the array elements and scanning radius on the object amplitude is studied. The final ratio of the object amplitude to the random noise level in the case of static accumulation and circular microscanning in the start/stop mode are compared for arrays with different sizes of the gap between the array elements.

This papers describes the effectiveness of textural features on the basis of gradient tensors in the problem of segmentation of images containing textures with structural redundancy. It is shown that the use of the bank of Gaussian bandpass filters in the formation of a feature space which is invariant to shifts and rotations of textures improves the performance of the segmentation algorithm. Examples demonstrating the application of textural features to the problem of segmentation with an unknown number of texture classes contained in the image are presented.

A new goodness-of-fit test constructed with the use of the interval estimation of the probability distribution function is proposed. A comparative analysis of this test with the Kolmogorov test is carried out. It is shown that it is possible to select the correct distribution function and the hypothetical function different from it so that the proposed goodness-of-fit test can have a significantly better performance than the Kolmogorov test. The presented numerical simulation results are in good agreement with the theoretical conclusions.

With regard to the problems of passive ranging, one of the modifications of the wellknown energy method for determining the range to an emitting target with partially known motion parameters on the basis of a single receiving point is developed. The problem is solved using solely the measurements of the power (amplitude) of the received signal on a predetermined time grid with no account for direction-finding data. The analytical dependences allowing for exploration of the working region of the method are obtained. An algorithm for the formation of the resulting ranging estimate in the current time taking into consideration the measurement redundancy is constructed. The results of computational experiments that confirm the possibility of practical implementation of the proposed modifications of the energy method are presented. The recommendations on the use of the results obtained in multipoint systems of passive location are given.

A problem of tracking of a group of identical objects with a variable shape and a large number of occlusions is considered. This problem is important for biology and safety systems. An algorithm of detection and tracking of biological objects on the basis of a sequence of depth images is proposed. Being tested, the algorithm provides one mistake per 193 collisions for the worst sequence.

The problem of face recognition is discussed. For this purpose, a method based on scalar perturbation functions and a set-theoretic operation of subtraction is proposed. It is shown that all surface points and the mask volume are used in the process of sample testing for more accurate identification.

A method has been developed to determine the shrinkage parameters of holographic photopolymer materials (including changes in the mean refractive index and thickness of the recording layer) during hologram recording with a measurement error acceptable for practical applications of these materials. Experimental verification of the method has proved its efficiency.

It has been shown that the problem of optimal actuator placement for adaptive square mirrors can be reduced to the analytical solution using elements of the theory of unbiased designs for the natural vibrations of a square plate with free edges. A stochastic model for the wavefront has been developed using regression analysis. Formulas are given for calculating the bias in working-out the desired optical surface geometry of an adaptive mirror.

A simple positioner, which allows one to ensure the position of a focused light beam on an object with nanometer precision and reproducibility, is proposed and studied. It can be used for positioning of focused beams (in particular, laser beams) on surfaces of optical fibers, biological objects, optical disks, thin film modulators, and holographic memory systems in the micrometer and nanometer ranges.

A simple method is proposed to determine the quality of antirelaxation coatings of the walls of vacuum optical cells containing alkali metal vapors. The method involves recording the time dependence of the fluorescence intensity of alkali atoms during exposure of the cell by resonant radiation pulses. Theory predicts that to determine the coating quality, it is sufficient to know the fluorescence decay rate and the ratio of the fluorescence intensities at the beginning and end of the pulse laser.

This paper considers a thin plane nanosystem consisting of molecular chains of dye J-aggregates forming bundles elongated in one direction and spherical gold nanoparticles located near them. Expressions are derived for the polarization of the exciton transition in the external field under the mutual influence of the dipole fields of the monomers constituting the dye J-aggregates and for the induced dipole of the nanoparticle. It is shown that in this system, mirrorless optical bistability of the fluorescence spectral density of the dye J-aggregates is possible, depending on the pump field intensity. The parameters of this dependence are critically determined by the distance between the molecular bundles and nanoparticles and by the angle between the field polarization direction and the axis of the bundle of J-aggregates. The use of the nanosystem as a sensor is discussed.

A new approach to constructing the thermal element of a multi-element array on the basis of thin pyroelectric films is developed: the element is neither thermally insulated from the substrate nor separated by a gap from the substrate surface; instead, it operates owing to accumulation of the charge generated by infrared radiation during the frame. A new principle of high-speed modulation of radiation is described. Experimental results for the behavior of the pyroelectric coefficient in barium-strontium niobate films (Ba_xSr_1-xNb₂O₆), whose values reach up to (1-2)·10^-3 C/m², are presented. It is demonstrated that the specific detectivity of sensitive elements 12 × 12 μm in size in the charge accumulation mode is greater than 10⁹ cm × Hz^0.5 ×·W^-1.

A method is proposed to shape the edges of chips in order to reduce the gap between chips in infrared-sensitive mosaic photodetectors modules. The method involves laser scribing of asymmetric grooves, splitting-off of the edge of the chip surface going under the chip, and the subsequent vertical alignment of the edges using laser radiation. The width of the gap between chips is 1-3 μm.

This paper describes the structure, operation algorithm, and circuitry of the recording channel of a multi-channel coordinate X-ray detector with a microsecond time resolution for studying the dynamics of density distribution in objects under pulsed loading. A method for fast automatic real-time control of the channel sensitivity is proposed.