Handwritten Digit Recognition Method Based on Sparse Transformation

A digital recognition and sparse transformation technology, applied in the field of pattern recognition, can solve problems such as data information redundancy, and achieve the effects of simple calculation, good recognition rate, and improved operating efficiency

Inactive Publication Date: 2018-07-03
XIAN UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a handwritten digit recognition method based on sparse transformation, which solves the problem of redundant data information caused by the digit adoption process existing in the prior art

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 Digit Recognition Method Based on Sparse Transformation
  • Handwritten Digit Recognition Method Based on Sparse Transformation
  • Handwritten Digit Recognition Method Based on Sparse Transformation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0047] The present invention is based on the handwritten digit recognition method of sparse transformation, adopts USPS database, and this database is the handwritten postal code that extracts from the mail envelope of the United States, altogether ten categories. The image size is 16×16, and the gray value range of each pixel is 0-255, such as figure 1 shown. The database includes 5,000 training data and 2,007 test data. The training data is arranged by category, and the test data is randomly arranged. The distribution of the number of samples contained in each category is shown in Table 1:

[0048] Table 1 Distribution of the number of samples contained in each type of training data and test data

[0049]

[0050] Select 2000 data from the training data for training (200 data for each category, 10 categories in total), and rando...

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 sparse transform-based handwritten digit recognition method. The method includes the following steps of: pre-processing read training data; training the pre-processed training data with an objective function so as to obtain a transform matrix W; performing sparse transform on data to be detected to obtain a sparse representation coefficient x'; and based on a nearest neighbor rule, comparing the sparse representation coefficient x' of the data to be detected with various classifications of class tags, and finding a class tag most similar to a sample to be detected, and classifying the sample to be detected to a classification to which the class tag belongs, and finishing recognition of handwritten digits. With the sparse transform-based handwritten digit recognition method of the invention adopted, after being subjected to the sparse transform, the data to be detected has sparsity, and therefore, computation is simple, and operation efficiency can be improved.

Description

technical field [0001] The invention belongs to the technical field of pattern recognition, and in particular relates to a method for recognizing handwritten digits based on sparse transformation. Background technique [0002] With the promotion of information network, there is a large amount of data to be input into computer network. Moreover, in the modern information society, all aspects must deal with numbers. As an important branch of pattern recognition, handwritten digit recognition has a wide range of applications in postal, taxation, transportation, financial and other industries. Not only that, handwritten digit recognition is used in some large-scale data statistics, such as industry annual inspection, population It can also be applied to fields such as censuses that require a lot of manpower and material resources. [0003] For identification or classification, data representation needs to highlight salient features, but in practice, the digital sampling proces...

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 Patents(China)
IPC IPC(8): G06K9/68
Inventor 梁军利叶欣贾薇李敏范文于国阳柯婷
Owner XIAN 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