ESN neural network image classification processing method based on memristor

A neural network and processing method technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as the stability of digital logic 0 and 1

Active Publication Date: 2020-08-18
NINGBO UNIVERSITY OF TECHNOLOGY
View PDF10 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0034] In addition, due to the problem of the memristor process, although the memristor can theor

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • ESN neural network image classification processing method based on memristor
  • ESN neural network image classification processing method based on memristor
  • ESN neural network image classification processing method based on memristor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0067] Attached below Figure 6 to 15 The realization of the present invention is described as follows:

[0068] The invention mainly uses the sandwich structure researched by the HP laboratory as the research object to construct the entire neural network model in hardware. According to the drift model, the corresponding circuit model can be obtained.

[0069] In the memristor circuit Indicates that there is a certain mapping relationship between magnetic flux and electric charge. Differentiate both ends at the same time with respect to time t to obtain formula (1):

[0070]

[0071] From the basic knowledge of the circuit, the relationship between the voltage and current across the memristor at time t is:

[0072]

[0073] Among them, M(q) is the resistance value of the memristor. So the relationship between memristance and the charge passed is:

[0074]

[0075] From the establishment of the physical model, memristance is mainly related to the width w(t) of TiO2-x, while the width ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses an ESN neural network image classification processing method based on a memristor, and relates to the technical field of image processing. The unique memory characteristics andoperational capability of the memristor are utilized, an echo state network is given, an ESN neural network circuit based on the memristor is designed to meet the requirement for the storage capability in the image processing process, memory access operation of training data is reduced, and finally the purpose of improving the performance and efficiency of the overall neural network is achieved.According to the method, data storage and operation based on the memristor are fused; image data is used as a training object; an image preprocessing function is realized by utilizing convolution operation of the image. According to the method, basic logic operations required by image preprocessing are screened out, circuit design of the memristor is performed on the basic logic operations by referring to implicit circuits, so that a data storage and operation structure based on the memristor is completed, and memory access operation of training data is reduced by combining storage and operation of image data. The application of the invention can improve the performance of the whole neural network.

Description

Technical field [0001] The invention relates to an ESN neural network image processing method, and relates to the technical field of image processing. Background technique [0002] With the advent of the era of big data, neural networks have shown good effects in solving many practical problems, such as image recognition, speech recognition and some other time series predictions. However, as the scale of data to be processed increases and the complexity of types increases, the problem of neural network performance becomes more and more prominent, and more time-sensitive neural network technology needs to be studied. [0003] As a new type of electronic device, the memristor has a unique memory function and non-volatile storage capacity, and has huge application potential in artificial neural networks. It is precisely because of these characteristics that it is hopeful to solve a series of problems in hardware implementation of neural networks. However, since the memristor is stil...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/62G06N3/04G06N3/063G06N3/08
CPCG06N3/063G06N3/08G06N3/045G06F18/24
Inventor 王建民贾仲言
Owner NINGBO UNIVERSITY OF TECHNOLOGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products