DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGET
Background. High accuracy of recognition of typical ground objects by optoelectronic tracking systems can be achieved by optimizing the parameters of an artificial neural network (INS) such as: the dimension and structure of the INS input signal, synapses of network neurons, the number of neurons...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Penza State University Publishing House
2022-02-01
|
| Series: | Надежность и качество сложных систем |
| Subjects: |
| _version_ | 1811303797334474752 |
|---|---|
| author | A.I. Godunov S.V. Shishkov S.T. Balanyan F.Kh. Al' Saftli |
| author_facet | A.I. Godunov S.V. Shishkov S.T. Balanyan F.Kh. Al' Saftli |
| author_sort | A.I. Godunov |
| collection | DOAJ |
| description | Background. High accuracy of recognition of typical ground objects by optoelectronic tracking systems
can be achieved by optimizing the parameters of an artificial neural network (INS) such as: the dimension and
structure of the INS input signal, synapses of network neurons, the number of neurons of each network layer and
the number of network layers. Materials and methods. The existing algorithms for optimizing the training of the
INS are considered when determining the number of neurons in the input, hidden and output layers of the INS in
order to increase the probability of recognizing images of a ground target. The factors of improving the training of
the INS, determining the number of neurons in the hidden layer for recognizing images of ground objects in such
algorithms as the Levenberg – Marquardt algorithm, the Bayesian regularization algorithm, the scalable conjugate gradient algorithm and the developed algorithm are investigated. Results and conclusions. The possibility of using
the developed algorithm in the subsystem of information and missile control during television homing on the target
is investigated. The software implementation of the developed algorithm using the Matlab programming language
is carried out. |
| first_indexed | 2024-04-13T07:54:34Z |
| format | Article |
| id | doaj.art-316413583a8f47dcab143a28c8d8c763 |
| institution | Directory Open Access Journal |
| issn | 2307-4205 |
| language | English |
| last_indexed | 2024-04-13T07:54:34Z |
| publishDate | 2022-02-01 |
| publisher | Penza State University Publishing House |
| record_format | Article |
| series | Надежность и качество сложных систем |
| spelling | doaj.art-316413583a8f47dcab143a28c8d8c7632022-12-22T02:55:26ZengPenza State University Publishing HouseНадежность и качество сложных систем2307-42052022-02-01410.21685/2307-4205-2021-4-12DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGETA.I. Godunov0S.V. Shishkov1S.T. Balanyan2F.Kh. Al' Saftli3Penza State UniversityBranch of the Military Academy of Logistics named after Army General A.V. Khrulev in PenzaAir Force Academy named after Professor N. E. Zhukovsky and Yu. A. GagarinAir Force Academy named after Professor N. E. Zhukovsky and Yu. A. GagarinBackground. High accuracy of recognition of typical ground objects by optoelectronic tracking systems can be achieved by optimizing the parameters of an artificial neural network (INS) such as: the dimension and structure of the INS input signal, synapses of network neurons, the number of neurons of each network layer and the number of network layers. Materials and methods. The existing algorithms for optimizing the training of the INS are considered when determining the number of neurons in the input, hidden and output layers of the INS in order to increase the probability of recognizing images of a ground target. The factors of improving the training of the INS, determining the number of neurons in the hidden layer for recognizing images of ground objects in such algorithms as the Levenberg – Marquardt algorithm, the Bayesian regularization algorithm, the scalable conjugate gradient algorithm and the developed algorithm are investigated. Results and conclusions. The possibility of using the developed algorithm in the subsystem of information and missile control during television homing on the target is investigated. The software implementation of the developed algorithm using the Matlab programming language is carried out.optimizationneural networkhidden layerneural network traininglevenberg – marquardt algorithmbayesian regularization algorithmscalable conjugate gradient algorithmrecognitionprobabilitygoal |
| spellingShingle | A.I. Godunov S.V. Shishkov S.T. Balanyan F.Kh. Al' Saftli DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGET Надежность и качество сложных систем optimization neural network hidden layer neural network training levenberg – marquardt algorithm bayesian regularization algorithm scalable conjugate gradient algorithm recognition probability goal |
| title | DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGET |
| title_full | DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGET |
| title_fullStr | DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGET |
| title_full_unstemmed | DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGET |
| title_short | DEVELOPMENT OF AN ALGORITHM FOR OPTIMIZING NEURAL NETWORK TRAINING WHEN DETERMINING THE NUMBER OF NEURONS IN A HIDDEN LAYER IN ORDER TO INCREASE THE PROBABILITY OF RECOGNIZING IMAGES OF A GROUND TARGET |
| title_sort | development of an algorithm for optimizing neural network training when determining the number of neurons in a hidden layer in order to increase the probability of recognizing images of a ground target |
| topic | optimization neural network hidden layer neural network training levenberg – marquardt algorithm bayesian regularization algorithm scalable conjugate gradient algorithm recognition probability goal |
| work_keys_str_mv | AT aigodunov developmentofanalgorithmforoptimizingneuralnetworktrainingwhendeterminingthenumberofneuronsinahiddenlayerinordertoincreasetheprobabilityofrecognizingimagesofagroundtarget AT svshishkov developmentofanalgorithmforoptimizingneuralnetworktrainingwhendeterminingthenumberofneuronsinahiddenlayerinordertoincreasetheprobabilityofrecognizingimagesofagroundtarget AT stbalanyan developmentofanalgorithmforoptimizingneuralnetworktrainingwhendeterminingthenumberofneuronsinahiddenlayerinordertoincreasetheprobabilityofrecognizingimagesofagroundtarget AT fkhalsaftli developmentofanalgorithmforoptimizingneuralnetworktrainingwhendeterminingthenumberofneuronsinahiddenlayerinordertoincreasetheprobabilityofrecognizingimagesofagroundtarget |