The structures of optical neural nets based on new matrix-tensor equivalental models (MTEMs) and results of modeling

Krasilenko, Vladimir G.; Nikolskyy, Aleksandr I.; Flavitskaya, J. A.

doi:10.3103/s1060992x10010054

Cited by 8 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These signals are actually coefficients β in appropriate models described in Section 2.1. If the models use only weighing by this coefficient β of appropriate weighting matrix or tensor synapse connections, as it has been shown in papers 5,10,12,16 , increasing the k-order of nonlinear transformations are necessary for improving adaptation and increasing the quality of the similarity measure determination of images. If the models use the additional weighing of matrices or tensors connections array of coefficients α (vector for vector or matrix for 2D input data) according to formula (1), in architecture, as seen in Fig.…”

Section: Adapting Optical Multiport Architecture Of Associative Memormentioning

confidence: 97%

“…This can be achieved only with corresponding architectures based on optical and optoelectronic realization with spatial and time integration 10,12 . In paper 16 it was shown that the optoelectronic architecture with time integration can provide significant benefits and performance 10 9 -10 10 connections per second and they can be used to realize and multiport autoassociative and multiport heteroassociative memory (MNNAAM and MNNHAM). The .…”

Section: Adapting Optical Multiport Architecture Of Associative Memormentioning

confidence: 99%

“…In papers 5,12,15,16 it has been shown that the implementation of the MME is possible on the basis of two matrix-matrix multipliers (MMM).…”

Section: The Ems Of Mham and Their Adaptation For Cluster Analysismentioning

confidence: 99%

“…At the same time, knowing the significant advantages of equivalence models (EM) when creating on their basis improved neural network recognition systems [7][8][9][10][11][12] and multiport autoassociative and heteroassociative memory [10][11][12] , there was a suggestion about the possibility of modifying EM and MHAM for parallel cluster image analysis [13][14] . Hardware implementations of these models are based on optoelectronic structures, including matrix-tensor multipliers, equivalentors and correlators with spatial and tine integration 15,16 . But in papers 14 -16 models and their implementation for recognition and clustering of images without spatial displacements, ie spatially non-invariant models were considered In paper 17 space-invariant models for image recognition, but not their clustering are examined Therefore abovementioned review and analysis of publications allowed us to determine the purpose of this paper, which is to adapt and modify models of MAM based on equivalency paradigm and architectures on a spatially invariant models for parallel cluster image analysis, modeling processes of such modified image clusterization, assessing performance and quality of these processes and the study of possible ways of learning and selflearning in such implementations.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Experimental research of methods for clustering and selecting image fragments using spatial invariant equivalent models

Krasilenko¹,

Lazarev

Nikitovich³

2014

SPIE Proceedings

View full text Add to dashboard Cite

In the paper, we show that the nonlinear spatial non-linear equivalency functions on the basis of continuous logic equivalence (nonequivalence) operations have better discriminatory properties for comparing images. Further, using the equivalent model of multiport neural networks and associative memory, (including matrix-matrix and matrix-tensor with adaptive-weighted correlation, multi-port neural-net auto-associative and hetero-associative memory (MP NN AAM and HAM ) and the proposed architecture based on them, we show how we can modify these models and architectures for space-invariant associative recognition and clustering (high performance parallel clustering processing) images. We consider possible implementations of 2D image classifiers, devices for partitioning image fragments into clusters and their architectures. The main base unit of such architectures is a matrix-matrix or matrix-tensor equivalentor, which can be implemented on the basis of two traditional correlators. We show that the classifiers based on the equivalency paradigm and optoelectronic architectures with space-time integration and parallel-serial 2D images processing have advantages such as increased memory capacity (more than ten times of the number of neurons!), High performance in different modes . We present the results of associative significant dimension (128x128, 610x340) image recognitionrenewal modeling. It will be shown that these models are capable to recognize images with a significant percentage (20-30%) damaged pixels. The experimental results show that such models can be successfully used for auto-and heteroassociative pattern recognition. We show simulation results of using these modifications for clustering and learning models and algorithms for cluster analysis of specific images and divide them into categories of the array. Show example of a cluster division of image fragments, letters and graphics for clusters with simultaneous formation of the outputweighted spatial allocated images for each cluster. Show results of other of modeling experiments with images of large dimension, such as clustering fragments (blocks 7x 7, 3x3, 15x15 and other sizes), 610x340 elements images into 8 clusters. We show that it is the use of nonlinear processing and nonlinear functions improves the quality of classification and image recognition. We offer criteria for the quality evaluation of patterns clustering with such MP NN AAM. It is shown that time of learning in the proposed structures of multi-port neural net classifier / categorizer-clustering (MP NN C) on the basis of equivalency paradigm, due to their multi-port, decreases by orders and can be, in some cases, just a few epochs. Other experimental data is also shown.

show abstract

Section: Adapting Optical Multiport Architecture Of Associative Memormentioning

confidence: 97%

Section: Adapting Optical Multiport Architecture Of Associative Memormentioning

confidence: 99%

“…In papers 5,12,15,16 it has been shown that the implementation of the MME is possible on the basis of two matrix-matrix multipliers (MMM).…”

Section: The Ems Of Mham and Their Adaptation For Cluster Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Experimental research of methods for clustering and selecting image fragments using spatial invariant equivalent models

Krasilenko¹,

Lazarev

Nikitovich³

2014

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…The works [6][7][8][9][10][11][12] , solve the problem of increase of capacity in NN and AM, even at storing of greatly correlated patterns, and the problem the convergence of methods and training rules, using multilevel representation of signals. The use of operations of neural logic-equivalence and nonequivalence for building of ANN and AM models is common for works [6][7][8][9][10][11][12][13] . In this connection such models of ANN and AM and the theory were called "equivalently".…”

Section: Introductionmentioning

confidence: 99%

The concept models and implementations of multiport neural net associative memory for 2D patterns

Krasilenko

Nikolskyy

Yatskovskaya

et al. 2011

Optical Pattern Recognition XXII

View full text Add to dashboard Cite

The paper considers neural net models and training and recognizing algorithms with base neurobiologic operations: p-step autoequivalence and non-equivalenc The Modified equivalently models (MEMs) of multiport neural net associative memory (MNNAM) are offered with double adaptive -equivalently weighing (DAEW) for recognition of 2D-patterns (images). It is shown, the computing process in MNNAM under using the proposed MEMs, is reduced to two-step and multi-step algorithms and step-by-step matrix-matrix (tensor-tensor) procedures. The given results of computer simulations confirmed the perspective of such models. Besides the result was received when MNNAM capacity on base of MEMs exceeded the amount of neurons.

show abstract

Design and Simulation of Array Cells of Mixed Sensor Processors for Intensity Transformation and Analog-Digital Coding in Machine Vision

Krasilenko

Lazarev

Nikitovich

2019

Machine Vision and Navigation

View full text Add to dashboard Cite

The structures of optical neural nets based on new matrix-tensor equivalental models (MTEMs) and results of modeling

Cited by 8 publications

References 8 publications

Experimental research of methods for clustering and selecting image fragments using spatial invariant equivalent models

Experimental research of methods for clustering and selecting image fragments using spatial invariant equivalent models

The concept models and implementations of multiport neural net associative memory for 2D patterns

Design and Simulation of Array Cells of Mixed Sensor Processors for Intensity Transformation and Analog-Digital Coding in Machine Vision

Contact Info

Product

Resources

About