Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Real-time handwritten digital recognition method based on multi-feature fusion

A multi-feature fusion and digital recognition technology, which is applied in character and pattern recognition, instruments, computer parts, etc., can solve problems such as difficult to achieve high-precision effective recognition, and achieve the effects of easy distinction, improved recognition rate, and obvious characteristics

Active Publication Date: 2014-08-20
WUHAN UNIV OF SCI & TECH
View PDF3 Cites 30 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] For handwritten digit recognition, it is difficult to achieve effective recognition with high precision

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
  • Real-time handwritten digital recognition method based on multi-feature fusion
  • Real-time handwritten digital recognition method based on multi-feature fusion
  • Real-time handwritten digital recognition method based on multi-feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0065] A real-time handwritten digit recognition method based on multi-feature fusion. The method as figure 1 As shown, the specific steps are as follows:

[0066] The first step, preprocessing of handwritten digital images

[0067] Preprocessing the original handwritten digital image I, including black and white binarization, intercepting the digital part, image adjustment, normalization, and thinning, to obtain the preprocessed image I 4 .

[0068] (1) The handwritten digital image I is converted into a bitmap format of 256 gray levels, and black and white binary processing is carried out, as follows:

[0069] f ( x , y ) = 0 , if f ( x , y ) ...

Embodiment 2

[0115] A real-time handwritten digit recognition method based on multi-feature fusion. as attached figure 2 As shown, taking the handwritten digital image "8" as an example of recognition, real-time recognition based on multi-feature fusion is carried out. The specific steps are as follows:

[0116] The first step, preprocessing of handwritten digital images

[0117] Preprocessing the original handwritten digital image I, including black and white binarization, intercepting the digital part, image adjustment, normalization, and thinning, to obtain the preprocessed image I 4 .

[0118] (1) Convert the handwritten digital image "8" into a bitmap format of 256 gray levels, and then set the gray value of all pixels whose gray value is less than 110 in the handwritten digital image "8" to 0, and the remaining pixels The gray value of the point is set to 255, and replaced with a binary image, using I 1 express;

[0119] (2) Intercept binary image I 1 The part of the handwritt...

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 real-time handwritten digital recognition method based on multi-feature fusion. Firstly, images in a handwritten digital image database are preprocessed, wherein the preprocessing steps comprise black and white binarization, digital part intercepting, image adjusting, normalization and refinement; then structural features and statistical features of the preprocessed images are extracted and fused, and a feature vector set is obtained; training learning is carried out by the utilization of a back-propagation neural network. The real-time handwritten digital recognition method based on multi-feature fusion not only reserves the authentication information in the structural features and the statistical features, but also eliminates redundant information to a certain degree, so that the features of all handwritten digital categories are more remarkable, the handwritten digital categories are prone to be distinguished, and a better recognition result is obtained.

Description

technical field [0001] The invention relates to the technical field of handwritten digit recognition, in particular to a real-time handwritten digit recognition method based on multi-feature fusion. Background technique [0002] Handwritten digit recognition is a traditional and typical pattern recognition problem, an important part of Optical Character Recognition (OCR), and has a wide range of applications in real life. Because handwritten digital images lack the dynamic information of strokes, and the writing styles are different, it is very difficult to recognize in the process of recognition, and the misrecognition rate is also high. However, handwritten digit recognition plays an irreplaceable role in some special occasions, such as postal code recognition in postal letter sorting, handwritten digit recognition in bank checks, etc. Since handwritten digit recognition often involves accounting and financial fields, its rigor is self-evident. Therefore, the reliability ...

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): G06K9/80
Inventor 张鸿马彩云
Owner WUHAN UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com