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

Optical character recognition method based on LVQ neural network

A technology of optical character recognition and neural network, applied in the field of intelligent recognition of small character sets, to achieve dynamic customization, reduce error rate, and strong learning and self-adaptive capabilities

Inactive Publication Date: 2014-04-23
UNIT 63680 OF PLA
View PDF1 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is currently no small character set intelligent recognition method and related products that have strict requirements on the error rate for the field of defense science and technology, such as the identification and verification of remote control code tables, spacecraft ballistics, and other special documents.

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
  • Optical character recognition method based on LVQ neural network
  • Optical character recognition method based on LVQ neural network
  • Optical character recognition method based on LVQ neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0034] Implementation Mode 1: see figure 1 As shown, the optical character recognition method based on the LVQ artificial neural network described in the present embodiment, its specific process is as follows:

[0035]Step A, preprocessing the image of the character to be recognized, including image denoising, character segmentation, binarization and feature extraction. Among them, the image denoising process is to remove noise signals such as ink dots and creases generated during the sampling and data transmission of the character image; the character segmentation process is to divide the entire character image into rows and columns, and obtain the size and position of each character; Binarization processing can reduce a large amount of redundant information, and convert a single character image into a two-dimensional pixel matrix, 0 represents white pixels, and 1 represents black pixels; the feature extraction process is to perform certain operations on the two-dimensional ...

Embodiment approach 2

[0047] Embodiment 2: Referring to Embodiment 1, the difference is that in step C, the similarity between each competitive layer neuron and the input vector is calculated according to formula (6), and the competitive layer neuron with the largest similarity is the winning neuron neuron, denoted as K1; the second winning neuron, denoted as K2.

[0048] (6)

[0049] Correspondingly, in step D, check the winning neuron, if and , then mark the character as a non-target character and refuse recognition; otherwise, go to step E. in is the minimum acceptable similarity.

Embodiment approach 3

[0050] Embodiment 3: Referring to specific embodiments 1 and 2, the difference is that in step E, the learning rate With the number of training times adjusted synchronously according to formula (7):

[0051] (7)

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 relates to an optical character recognition method based on an LVQ (Learning Vector Quantization) neural network. The method includes the following steps: performing preprocessing including image denoising, character segmentation, binaryzation and characteristic extraction on a to-be-recognized character image; sending a single character characteristic into an input layer of the LVQ neural network and acquiring a winning competition-layer nerve cell according to a competition algorithm and performing a recognition rejection test; and inputting a character characteristic which complies with a recognition condition to the competition layer of the LVQ neural network for recognition and outputting the recognition result. The optical character recognition method based on the LVQ neural network is applicable to the field of automatic recognition of a small character set which has strict demands on false rate.

Description

technical field [0001] The invention relates to an optical character recognition method based on learning vector quantization (LVQ, Learning Vector Quantization) neural network, which is mainly used for intelligent recognition of small character sets with strict requirements on misrecognition rate. Background technique [0002] Optical Character Recognition (Optical Character Recognition, OCR) obtains text image information on paper through optical input methods such as scanning and photography, uses various pattern recognition algorithms to analyze text morphological features, and determines the standard code of text, and stores them in a common format. in the text file. In the past few decades, OCR has become the research goal of many researchers, and it has important implications in many different fields, such as banking, shipping, commerce, communication, and national defense. [0003] German scientist Tausheck formally proposed the concept of OCR in 1929, but it was no...

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): G06K9/20G06K9/60G06N3/02
Inventor 张煜昕李永刚何剑伟周琳琦郭力兵李祥明茅文浩李清梅胡上成张龙
Owner UNIT 63680 OF PLA
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