Lao language printed text optical character recognition method based on convolutional neural network

A convolutional neural network and optical character recognition technology, applied in character and pattern recognition, instruments, computer components, etc.

Pending Publication Date: 2019-07-09
KUNMING UNIV OF SCI & TECH
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a method for optical character recognition of Lao printed text based on convolutional neural network, which is used to solve the problem of digital entry of Lao paper materials

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
  • Lao language printed text optical character recognition method based on convolutional neural network
  • Lao language printed text optical character recognition method based on convolutional neural network
  • Lao language printed text optical character recognition method based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] Embodiment 1: as Figure 1-5 As shown, a method for optical character recognition of Lao printed text based on convolutional neural network includes the following steps:

[0025] Step1, input the digital image that can be processed by the computer, and the input image is the scanned picture of the printed matter scanned by the scanner. Specifically, the present embodiment uses the Python language, and reads the printed matter scan picture from the local disk by using the open CV library, and stores it in the memory in the form of an image;

[0026] Step2, preprocessing the image, that is, through digital image processing technology, the input image is converted into a binary black and white image according to the local adaptive binarization method, noise is eliminated, and then the image distortion is corrected according to the minimum circumscribed rectangle.

[0027] Step3, analyze the binarized undistorted image first horizontally and then vertically by the projecti...

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 Lao language printed text optical character recognition method based on a convolutional neural network, and belongs to the technical field of natural language processing andmachine learning. According to the method, after a printed text is input to scan a picture, binarization processing is firstly carried out on the image, and rotation correction is carried out. Then, Laos are horizontal vowel attached label characters written from left to right, so that the whole page of characters are cut into Laos character combinations according to a sequence of preceding and following by using a projection histogram method on the preprocessed image. Next, the segmented Lao characters are input into a four-way parallel convolutional neural network model established for Lao language characteristics, and outputting corresponding character texts. Finally, post-processing operation is carried out on the character text sequence according to Lao language rules to generate final text output. The Lao character recognition method has a certain application value in Lao character recognition and paper data digital processing.

Description

technical field [0001] The invention relates to an optical character recognition method for Lao printed text based on a convolutional neural network, which belongs to the field of optical character recognition for small and medium-sized languages ​​in natural language processing, and is applicable to the application scene of Lao image-to-text conversion. Background technique [0002] Optical character recognition is to use computer to automatically recognize characters printed on paper or written by people on paper. It is an important branch of pattern recognition and natural language processing. At present, the research on optical character recognition in the world is mainly aimed at the recognition of mainstream languages ​​such as Chinese and English. There are also some researches on the recognition of minority characters in China, mainly Mongolian, Tibetan, Uighur and other characters. One of the main research problems of optical character recognition is the classificat...

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 & AuthorityApplications(China)
IPC IPC(8): G06K9/32G06K9/34
CPCG06V20/63G06V30/153
Inventor周兰江郝永彬周枫张建安
OwnerKUNMING UNIV OF SCI & TECH