Development of the modified methods to train a neural network to solve the task on recognition of road users

Abstract

<p>We have developed modifications of a simple genetic algorithm for pattern recognition. In the proposed modification Alpha-Beta, at the stage of selection of individuals to the new population the individuals are ranked in terms of fitness, then the number of pairs is randomly determined ‒ a certain number of the fittest individuals, and the same number of the least adapted. The fittest individuals form the subset B, those least adapted ‒ the subset W. Both subsets are included in a set of pairs V. The number of individuals that can be selected to pairs is in the range of 20‒60 % of the total number of individuals. In the modification Alpha Beta fixed compared to the original version of a simple genetic algorithm we added a possibility of the emergence of two mutations, added a fixed point of intersection, as well as changed the selection of individuals for crossbreeding. This makes it possible to increase the indicator of accuracy in comparison with the basic version of a simple genetic algorithm. In the modification Fixed a fixed point of intersection was established. The cross-breeding involves half the genes ‒ those genes that are responsible for the number of neurons in layers, values for other genes are always passed to the descendants from one of the individuals. In addition, at the stage of mutation there are randomly occurring mutations using a Monte-Carlo method.</p>The developed methods were implemented in software to solve the task on recognizing motorists (cars, bicycles, pedestrians, motorcycles, trucks). We also compared indicators for using modifications of a simple genetic algorithm and determined the best approach to solving the task on recognizing road traffic participants. It was found that the developed modification Alpha-Beta showed better results compared to other modifications when solving the task on recognizing road traffic participants. When applying the developed modifications, the following indicators for the accuracy of Alpha-Beta were obtained ‒ 96.90 %, Alpha‒Beta fixed ‒ 95.89 %, fixed ‒ 85.48 %. In addition, applying the developed modifications reduces the time for the neuromodel’s parameters selection, specifically using the Alpha-Beta modification employs only 73.9 % of the time required by the basic method, applying the Fixed modification ‒ 91.1 % of the time required by the basic genetic method

Authors and Affiliations

Ievgen Fedorchenko, Andrii Oliinyk, Alexander Stepanenko, Tetiana Zaiko, Serhii Shylo, Anton Svyrydenko

Keywords

Related Articles

The development of methods for determining vibration stochastic fields of technological complexes

<p>The force effects occurring in technological complexes have been studied on the basis of the analysis of technical diagnostics system. Due to the distinction between deterministic and random force effects, there have...

Electrophoretic system for express analysis of whey protein fractions

<p>Fractional specificity of biological action and ability to form bioactive peptides, which have a positive effect on different physiological systems of the body in the processes of proteolysis and digestion, are charac...

Mathematical model of obtaining a hydrocarbon fuel based on the fischer­tropsch pathway in a stationary layer of the cobalt­based catalyst

<p>Studying the kinetics of the Fischer-Tropsch process is a rather important task, since this process is very sensitive to the temperature mode, as well as to the characteristics of the catalyst. In addition, a given pr...

Computer simulation methods of redundant measurements with the nonlinear transformation function

<p>The studies of possibilities of methods of redundant measurements have found the high efficiency of the presented methods for increasing the accuracy of measurements. It is proved that the equation of redundant measur...

Influence of storage duration and density of raw cotton on the mechanics of the interaction process between feeding rollers in the cleaners of large impurities

<p>We have examined influence of storage duration and the density of cotton in bale on the adhesion between impurities and cotton. We found the effect of raw cotton storage, as well as the fiber, in a compacted state, wh...

Download PDF file
  • EP ID EP666325
  • DOI 10.15587/1729-4061.2019.164789
  • Views 98
  • Downloads 0

How To Cite

Ievgen Fedorchenko, Andrii Oliinyk, Alexander Stepanenko, Tetiana Zaiko, Serhii Shylo, Anton Svyrydenko (2019). Development of the modified methods to train a neural network to solve the task on recognition of road users. Восточно-Европейский журнал передовых технологий, 2(9), 46-55. https://europub.co.uk/articles/-A-666325