Unlock instant, AI-driven research and patent intelligence for your innovation.

Oracle bone character classification method based on graph convolutional network

A technology of convolutional network and classification method, which is applied in the field of oracle bone character classification based on graph convolutional network, can solve the problem of low prediction accuracy of difficult to distinguish variant oracle bone characters, and achieve the effect of improving the accuracy rate

Pending Publication Date: 2022-02-11
ANYANG NORMAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Classification of oracle bone characters with similar semantics has the problem of low prediction accuracy of difficult to distinguish variant oracle bone characters

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
  • Oracle bone character classification method based on graph convolutional network
  • Oracle bone character classification method based on graph convolutional network
  • Oracle bone character classification method based on graph convolutional network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0059] 1 Network construction

[0060] 1.1 Oracle Bone Character Context Network Construction

[0061] The inscriptions in the oracle bone rubbings describe a relatively complete scene and event. Therefore, the relationship between the contexts of the oracle bone characters in the same oracle bone tablet reflects the contextual information of the oracle bone characters. This paper takes the same rubbing as the context unit, quantifies the context information by defining the weight between any oracle bone characters in the same rubbing, and stores the oracle bone character context data in the same context network (graph). The specific steps are as follows. First, select a piece of rubbing O=[1,2,...,i,...,j,...,n] containing n oracle bone characters. Assuming that there are two oracle bone characters i and j on this rubbing, we The distance w between oracle bone characters i and j can be defined using formulas (1) and (2) ij , and put w ij Value assigned to M ij (M is the s...

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 oracle bone character classification method based on a graph convolutional network, and the method comprises the following steps of: firstly, constructing a context and font network of oracle bone characters, then, screening allogeneic oracle bone characters with large font difference and similar semantics, and extracting an oracle bone character sub-network with similar semantics, and finally, classifying the oracle bone characters with similar semantics by utilizing the graph convolutional network, and analyzing the effect of context and font information in the oracle bone character semantic forming process. The method can effectively improve the accuracy of semantic prediction.

Description

technical field [0001] The present invention relates to the technical field of oracle bone character classification, and more specifically relates to a method for classifying oracle bone characters based on a graph convolutional network. Background technique [0002] The method based on image processing is to recognize oracle bone characters from the overall macroscopic structure of oracle bone characters, while some literatures recognize oracle bone characters from the microscopic perspective of oracle bone characters' strokes or components (ie a part of oracle bone characters). In 2011, Li Qingsheng and others first abstracted some important endpoints and intersection points of oracle bone characters as points in the graph; then, connected the edges between the points in the graph according to the oracle bone characters to form the graph of the oracle bone characters; finally, Identify unknown oracle bone characters according to the matching algorithm of the graph. In 201...

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): G06V10/764G06K9/62G06N3/04G06F40/30
CPCG06F40/30G06N3/047G06N3/045G06F18/2415
Inventor 刘国英焦清局吴亮葛文英
Owner ANYANG NORMAL UNIV