Abstract
Large enough structured neural networks are used for solving the tasks to recognize distorted images involving computer systems. One such neural network that can completely restore a distorted image is a fully connected pseudospin (dipole) neural network that possesses associative memory. When submitting some image to its input, it automatically selects and outputs the image that is closest to the input one. This image is stored in the neural network memory within the Hopfield paradigm. Within this paradigm, it is possible to memorize and reproduce arrays of information that have their own internal structure. In order to reduce learning time, the size of the neural network is minimized by simplifying its structure based on one of the approaches: underlying the first is «regularization» while the second is based on the removal of synaptic connections from the neural network. In this work, the simplification of the structure of a fully connected dipole neural network is based on the dipole-dipole interaction between the nearest adjacent neurons of the network. It is proposed to minimize the size of a neural network through dipole-dipole synaptic connections between the nearest neurons, which reduces the time of the computational resource in the recognition of distorted images. The ratio for weight coefficients of synaptic connections between neurons in dipole approximation has been derived. A training algorithm has been built for a dipole neural network with sparse synaptic connections, which is based on the dipole-dipole interaction between the nearest neurons. A computer experiment was conducted that showed that the neural network with sparse dipole connections recognizes distorted images 3 times faster (numbers from 0 to 9, which are shown at 25 pixels), compared to a fully connected neural network
Highlights
Intensive research is underway on the use of neural networks to solve a wide class of problems related to intelligent data analysis
A computer experiment was conducted that showed that the neural network with sparse dipole connections recognizes distorted images 3 times faster, compared to a fully connected neural network
Study [4] analyzed the effects of white Gaussian noise, pixel maximization, and brightness minimization on image recognition quality; paper [5] examined the neural network technology of recognition of handwritten documents where the role of distorted images belongs to different styles of writing letters
Summary
Intensive research is underway on the use of neural networks to solve a wide class of problems related to intelligent data analysis (detection of non-stationary chaotic processes, clustering, intelligent control, diagnostics of biosystems, forecasting, emulation and pattern recognition). Study [4] analyzed the effects of white Gaussian noise, pixel maximization, and brightness minimization on image recognition quality; paper [5] examined the neural network technology of recognition of handwritten documents where the role of distorted images belongs to different styles of writing letters. It is a relevant task to design the architecture of an artificial neural network with synaptic connections between the nearest dipole neurons since such a neural network could improve the efficiency of recognizing distorted images by reducing the computing resource used. With increasing requirements for efficiency in solving problems of recognition of distorted images, we propose to create a neural network with sparse dipole connections based on a prototype of the natural neural network of the cytoske leton microtube type, which would reduce the computing resource used to recognize distorted images
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
