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

Self-learning-based handwritten form numeric character string rapid recognition method

A technology of digital characters and strings, which is applied in the field of rapid recognition of digital strings in handwritten forms based on self-learning, can solve the problem of low recognition rate of forms, and achieve the effect of convenient segmentation and recognition

Inactive Publication Date: 2015-12-23
HARBIN INST OF TECH
View PDF3 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem of low form recognition rate in existing methods, and proposes a method for fast recognition of handwritten form digital strings based on self-learning

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
  • Self-learning-based handwritten form numeric character string rapid recognition method
  • Self-learning-based handwritten form numeric character string rapid recognition method
  • Self-learning-based handwritten form numeric character string rapid recognition method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0027] Specific embodiment one: a kind of method based on self-study handwritten form numerical string rapid recognition, it is characterized in that, a kind of method based on self-study handwritten form numerical string fast recognition is specifically carried out according to the following steps:

[0028] Step 1, preprocessing the form image;

[0029] Step 2, extracting and segmenting the digital characters in the preprocessed form image;

[0030] Step 3: Recognize the extracted and segmented digital characters.

specific Embodiment approach 2

[0031] Specific embodiment 2. The difference between this embodiment and specific embodiment 1 is that the form image is preprocessed in the step 1; the specific process is:

[0032] Step 11, binarizing the table image;

[0033] Set 0 as the foreground (i.e. target) value is black, 255 as the background value is white, threshold T th Choose to maximize the between-class variance δ;

[0034] Step 12, denoising the form image;

[0035] The median filtering method was used to remove noise.

[0036] Other steps and parameters are the same as those in Embodiment 1.

specific Embodiment approach 3

[0037] Specific embodiment three, the difference between this embodiment and specific embodiment one or two is that in said step two, the digital characters in the preprocessed form image are extracted and segmented; the specific process is:

[0038] Step 21. Detect the row coordinates and column coordinates of each cell in the preprocessed table image, and locate the row coordinates and column coordinates of each cell in the detected preprocessed table image. The specific process is: cell The row coordinate of the cell is the horizontal line of the cell, and the column coordinate of the cell is the vertical line of the cell; Viterbi algorithm is used to detect the horizontal line and vertical line of the cell. The Viterbi algorithm is a dynamic programming algorithm used to find the most likely Observe the Viterbi path of the event sequence, detect the straight line (horizontal line and vertical line), that is, use each foreground point in the horizontal line and vertical line...

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 a self-learning-based handwritten form numeric character string rapid recognition method, and aims to solve the problem of low form recognition rate of the existing method. The self-learning-based handwritten form numeric character string rapid recognition method is implemented through the technical scheme of: step 1, preprocessing a form image; step 2, extracting and segmenting numeric characters in the preprocessed form image; step 3, and recognizing the numeric characters after extraction and segmentation. The self-learning-based handwritten form numeric character string rapid recognition method is applied to the field of form numeric character string recognition.

Description

technical field [0001] The invention relates to a method for fast recognition of handwritten form numeral strings based on self-learning. Background technique [0002] In daily life, people have to come into contact with digital tables every day, such as financial, performance statistics, experimental data, etc. The data processing of these table numbers is not only a heavy workload, but also boring. Therefore, if a method of automatically identifying scanned form documents is found and the data processing is handed over to the computer, time will be greatly saved and efficiency will be improved. If you want to identify the data of the table document, you must first process the structure of the table. First of all, the format of the table is ever-changing. At present, no method has been found that can be used in all tables. The recognition of the table structure still needs further in-depth research; secondly, although the format of the table is complex, it is often necessa...

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
IPC IPC(8): G06K9/00
CPCG06V30/40G06V30/10
Inventor 关宇东吴梦蝶朱瑞锋提纯利仲小挺
Owner HARBIN INST OF 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