Handwritten numeral recognition model training method and system

A technology for model training and digit recognition, which is applied in the field of handwritten digit recognition model training methods and systems, and can solve problems such as unfavorable promotion, unreasonable use of computer resources, and low training efficiency.

Active Publication Date: 2021-06-25
GUANGDONG UNIV OF TECH
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The invention provides a handwritten digit recognition model training method and system, which are used to solve the technical problems that the existing pulse neural network adopts a serial training method, cannot rationally utilize computer resources, has low training efficiency, and is not conducive to popularization

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
  • Handwritten numeral recognition model training method and system
  • Handwritten numeral recognition model training method and system
  • Handwritten numeral recognition model training method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0084] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0085] For ease of understanding, see Figure 1 to Figure 3 , the present invention provides an embodiment of a handwritten digit recognition model training method, comprising:

[0086] Step 101, input the MNIST training data set and the STDP synapse initial weight matrix into the global spiking neural network model, which includes each neuron m...

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 a handwritten numeral recognition model training method and system, and the method comprises the steps: inputting an MNIST training data set and an STDP synaptic initial weight matrix, building each neuron model and each synaptic model, employing a distributed multi-thread parallel technology, dynamically employing a plurality of threads to carry out the pre-division of a neuron group according to computer resources, and then establishing a local spiking neural network in the threads, initializing a neuron group, a synaptic connection relationship and a synaptic weight in each independent thread, and after initialization is completed, carrying out parallel training on all the threads according to a set number of iterations, thereby solving the problem that the existing spiking neural network adopts a serial training method, computer resources cannot be reasonably utilized, the training efficiency is low, and popularization is not facilitated.

Description

technical field [0001] The invention relates to the technical field of handwriting recognition, in particular to a handwritten number recognition model training method and system. Background technique [0002] Handwritten digit recognition is the ability of a computer to receive and understand and recognize readable handwritten digits from paper documents, photos or other sources. It has great practical application value in real life. For example, handwritten digit recognition can be applied in bank remittance In the single number recognition, to greatly reduce the labor cost, in this process, the deep artificial neural network structure is generally used for recognition. But the essence of deep artificial neural network is far from the actual brain model, and the recognition rate is not high. For this reason, the Spiking Neural Network (SNN), which uses the third-generation artificial neural network to improve and realize the information transmission between neurons in the...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04G06N3/08G06K9/62G06K9/00
CPCG06N3/049G06N3/08G06V30/32G06F18/214Y02D10/00
Inventor 林彦宇刘怡俊林文杰叶武剑刘文杰
Owner GUANGDONG UNIV OF TECH
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap