Handwriting number identification method and system

A technology of digital recognition and handwriting, applied in the field of analog recognition, can solve problems such as unbalanced sample distribution, inability to maintain local structures in the adjacency graph, poor classification performance, etc., and achieve the effect of overcoming poor classification performance

Inactive Publication Date: 2014-03-26
SUZHOU UNIV
View PDF6 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of this, this application provides a handwritten digit recognition method and system, which is used to solve the problem that the existing K-nearest neighbor classification algorithm is aimed at the situation of unbalanced sample distribution, and the adjacency graph constructed cannot well maintain the local structure, making the classification performance bad question

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
  • Handwriting number identification method and system
  • Handwriting number identification method and system
  • Handwriting number identification method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] see figure 1 , figure 1 It is a flow chart of a method for recognizing handwritten digits disclosed in the embodiment of this application.

[0057] Such as figure 1 As shown, the method includes:

[0058] Step 101: Construct an intra-class adjacency graph and an inter-class adjacency graph for the training samples according to the distance and the label category;

[0059] Specifically, the intra-class adjacency graph is to find the nearest neighbor samples among the same samples, and the inter-class adjacency graph is to find the nearest neighbor samples among the heterogeneous samples.

[0060] Step 102: Determine the target dimension and projection transformation matrix according to the matrix composed of the intra-class adjacency graph, the inter-class adjacency graph and the training samples;

[0061] Specifically, in order to simplify the calculation, we project the training samples into a space with relatively low dimensions. This process requires determining ...

Embodiment 2

[0069] In this embodiment, the above steps will be described in detail.

[0070] (1) Construct the intra-class adjacency graph and the inter-class adjacency graph for the training samples in the original space according to the distance and label category, so as to maintain the relationship between the neighbors.

[0071] Specifically, the training sample set is defined as

[0072] where y i is x i The category label of , c represents the number of categories, N represents the total number of training samples, and d represents the dimensionality of the training samples.

[0073] Define the intra-class adjacency graph as F w :

[0074]

[0075] in with Respectively represent the sample x i and x j The set of similar neighbors of the same class, and x i with x j The categories are the same. Define the class-to-class adjacency graph as F b :

[0076]

[0077] in with Respectively represent the sample x i and x j The heterogeneous neighbor set of , and ...

Embodiment 3

[0099] see figure 2 , figure 2 It is a structural diagram of a handwritten digit recognition system disclosed in the embodiment of this application.

[0100] Such as figure 2 As shown, the system includes:

[0101] The adjacency graph construction unit 21 is used for constructing the intra-class adjacency graph and the inter-class adjacency graph to the training sample according to the distance and the label category;

[0102] A projection transformation matrix determination unit 22, configured to determine the projection transformation matrix and the target dimension according to the matrix formed by the intra-class adjacency graph, the inter-class adjacency graph and the training samples;

[0103] A sample to be tested acquisition unit 23, configured to acquire a sample to be tested;

[0104] A sample transformation unit 24, configured to transform the training samples into the discriminant subspace according to the projection transformation matrix, and map the sample...

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 handwriting number identification method, which comprises the following steps that an intra-class abutting figure and an inter-class abutting figure are constructed on training samples according to the distance and the label types; a target dimension number and a projection conversion matrix are determined according to a matrix formed by the intra-class abutting figure, the inter-class abutting figure and the training samples; the training samples are converted in to a discrimination sub space according to the projection conversion matrix; the projection conversion matrix is utilized for mapping samples to be tested into the discrimination sub space to obtain test samples; a neighbor classifier is utilized for classifying the test samples. The method has the advantages that each sample is respectively subjected to neighbor finding in identical-class samples and neighbor finding in different-class samples, so the problem of poor classification performance caused by sample distribution unbalance is solved.

Description

technical field [0001] The present application relates to the technical field of analog recognition, and more specifically, to a method and system for recognizing handwritten digits. Background technique [0002] Handwritten digit recognition has always been a research hotspot in the field of analog recognition technology. In modern society, there are countless fields related to handwritten digit recognition, such as mail sorting, taxation, finance and other fields. With the development of the economy, there are more and more financial statements and checks waiting to be processed every day. If they can be processed automatically by computer, it will save a lot of financial, material and manpower. Therefore, the solution to this kind of problem is to design a reliable A handwritten digit recognition method with high accuracy and high recognition rate. [0003] In the prior art, one method is to use the K-nearest neighbor classifier, but the classifier must calculate its di...

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): G06K9/62
Inventor 张莉丁春涛严晨王邦军李凡长
Owner SUZHOU UNIV
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