The method of image restoration based on the mathematical apparatus of artificial neural networks of the advanced method of hierarchical interpolation

Abstract

The paper proposes a new compression method and an algorithm based on it, which is a generalization of the known method of decoding with transformation in the case of three-dimensional hyperspectral data. The task of choosing the parameters of the algorithm is solved. The research of algorithm in comparison with known methods of compression of images and hyperspectral data, its advantages are demonstrated. Among the known image compression methods, the most widely used method of encoding with transformation is implemented, in particular, in the standard (algorithm) of JPEG and the improved compression method (due to the mathematical apparatus of artificial neural networks) on the basis of hierarchical network interpolation. However, JPEG, although sometimes used to compress hyperspectral data, appears to be inefficient in this case. At the stage of restoration (decoding), a decoding and decantation procedure of generalized coordinates is performed. Then, for each block, the inverse transforms are calculated and the pixels of the image are restored.

Authors and Affiliations

A. Maziashvili, N. Korolyova, S. Indyk

Keywords

Related Articles

Energoefficiency analisys of active rectifier with improved hysteretic control system

The article describes the active three-phase boost rectifier with power factor correction, as known as active front-end rectifiers, with hysteretic control system in conditions of the ideal voltage and distorted voltage....

Calculation of the effect of nonlinear effects in DWDM systems on the energy characteristics of a signal

Nowadays DWDM (Dense Wavelength - Division Multiplexing) technology that allows increasing manyfold bandwidth capability of existing telecommunication systems and combining different types of telecommunication technologi...

Formation of the information and control system of strategic planning of speed passenger transportation on railway transport

An analysis of the implementation of recent projects to increase the speed of passenger trains in the existing system of managing rail passenger transportation in Ukraine allows us to draw the following conclusions: deci...

Thermal checking hot box system knots of rolling stock

One of the most responsible elements of construction of freight carriage, that directly influence on safety of motion, hot box knots are with bearing of that or other type. Therefore to development and perfection of appa...

Creating an effective technology for the formation railway routes with grain cargoes using models on the Petri net

The method of formalization technology of formation railway routes with grain cargoes with the models on Petri's nets is proposed. For the first time, modeling of railway technological processes of transportation in the...

Download PDF file
  • EP ID EP499732
  • DOI 10.18664/ikszt.v0i5.146695
  • Views 132
  • Downloads 0

How To Cite

A. Maziashvili, N. Korolyova, S. Indyk (2018). The method of image restoration based on the mathematical apparatus of artificial neural networks of the advanced method of hierarchical interpolation. Інформаційно-керуючі системи на залізничному транспорті , 23(5), 44-49. https://europub.co.uk/articles/-A-499732