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

Bill character recognition method and system based on convolutional neural network

A convolutional neural network and text recognition technology, applied in the field of neural networks, can solve the problems that the model is not well targeted and the suitability of Siemens invoices is not high, and achieves the effect of solving cumbersome processing.

Inactive Publication Date: 2020-12-11
CHANGSHU INSTITUTE OF TECHNOLOGY
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0017] The training models used by A and B are not well targeted, and the applicability to Siemens invoices is not high

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
  • Bill character recognition method and system based on convolutional neural network
  • Bill character recognition method and system based on convolutional neural network
  • Bill character recognition method and system based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0073] 1. Image preprocessing

[0074] Although the scanned trade bill looks like a black and white picture, it is actually an RGB image, which contains information of three channels: red (R), green (G), and blue (B). However, during the processing of such three-color channels, the amount of computation is huge. Therefore, using the OpenCV library that comes with Python, the image grayscale technology is first introduced to change the three channels into a single channel. Then, the gray value of each pixel in the image matrix is ​​changed to 0 (black) or 255 (white) by using binarization processing technology. In this way, the contrast of the picture is more obvious, only black and white.

[0075] 2. Field positioning and character segmentation

[0076] According to the analysis, each form has a fixed template, which means that the relative positions of each part are fixed. If you can find the position of a fixed mark on the picture, you can obtain the coordinates of the ke...

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 bill character recognition method and system based on a convolutional neural network, and the method comprises the steps: 1, carrying out the image preprocessing of an RGB image of a trade list, introducing image graying through an OpenCV library of Python, and enabling three channels to be changed into a single channel; adopting a binarization processing technology, so as to enable the gray value of each pixel point of the picture matrix to be changed into 0 or 255; step 2, carrying out field positioning and character segmentation, and determining and intercepting coordinate positions of key fields of the bill, wherein each form is provided with a fixed template; and if the position of a certain fixed identifier can be found on the picture, obtaining the coordinates of the key field through relative positioning; and step 3, performing training by adopting a convolutional neural network. According to the method, the convolutional neural network is adopted, a bill key field recognition system is achieved for Siemens trade invoices, the problem of tedious processing of traditional paper bills is effectively solved, and key fields of the bills can be automatically extracted and recognized.

Description

technical field [0001] The present invention relates to the field of neural networks, and more specifically relates to a bill text recognition method and system based on a convolutional neural network. Background technique [0002] OCR recognition technology (Optical Character Recognition) is the basis of bill text recognition, which is a technology that converts optical characters on pictures into editable text. The concept of OCR was first proposed by German scientists in the 1930s, and the first OCR character recognition software was able to recognize 120 English letters in one second. my country started relatively late in OCR technology research, and it was only in the 1970s that research on the recognition of numbers, English letters and symbols began. [0003] Over the years, with the advancement of OCR technology research, many domestic companies have begun to study text recognition technology in taxation, finance and other fields, and many professional companies hav...

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): G06K9/00G06K9/34G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V30/413G06V10/267G06V30/10G06N3/045G06F18/241
Inventor 徐江梁昊张金龙
Owner CHANGSHU INSTITUTE OF TECHNOLOGY
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