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

English handwriting identification method based on improved VGG-16 model

A VGG-16, English technology, applied in the field of computer vision, can solve the problems of imperfect handwriting identification theory, high degree of writing diversification, single structure, etc., and achieve the effect of rapid network stabilization, saving training time, and rapid convergence

Active Publication Date: 2019-08-13
TIANJIN UNIV
View PDF5 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The difficulty of English handwriting identification lies in: English characters belong to a foreign language handwriting, which has the characteristics of simple strokes, single structure, low writing difficulty, and high degree of writing diversity; at the same time, the theory of this type of handwriting identification is not perfect and the theoretical basis is insufficient. Scholars have done a lot of research work in this field, but there are still many deficiencies and lack of practicality

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
  • English handwriting identification method based on improved VGG-16 model
  • English handwriting identification method based on improved VGG-16 model
  • English handwriting identification method based on improved VGG-16 model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] A kind of English handwriting identification method based on improved VGG-16 model, the method comprises the following steps:

[0034] 101: Collect a dataset of English documents from different handwritings of different people;

[0035] 102: Segment the obtained handwritten English handwriting document into words to obtain a data set composed of English words, and construct a training sample set and a test sample set based on the data set;

[0036] 103: Based on the VGG-16 model, improve the VGG-16 model and construct a convolutional neural network. The improved VGG-16 model includes 2 traditional convolutional layers, 3 composite convolutional layers, 5 pooling layers and 3 fully connected layers;

[0037] 104: Input a training sample set, extract character features, and perform classification training; wherein, the sizes of the training sample set and the test sample set are both 320*320 pixels;

[0038] 105: Use the trained neural network to automatically identify ...

Embodiment 2

[0044] The following is combined with specific examples, calculation formulas, Figure 1-Figure 4 , Table 1 further introduces the scheme in Embodiment 1, see the following description for details:

[0045] 201: Obtain a training sample set and a test sample set for each English word;

[0046] First, the orthorectification is performed on the handwritten English document image, and then the single English word is segmented into rows and columns by the projection method to obtain the English word data set. Among them, the relevant texts of 130 people's handwritten English documents are collected as a data set, and each person writes an English document, including about 240 words, such as figure 1 As shown, classify tagged image information. Finally, the total number of words in the training sample set is 26,000, and the total number of words in the test sample set is 2,600. Normalize the size of the segmented English word image to 320*320 pixels, and set the data type to the...

Embodiment 3

[0073] Combine below Figure 5 , and Table 2-Table 4, the scheme in Embodiment 1 and 2 is further introduced, see the following description for details:

[0074] In order to verify the effectiveness of the method for English handwriting identification, it was tested, and the test results are as follows:

[0075] In the embodiment of the present invention, two evaluation standards, Mean Average Precision (MAP) and Hard top-k, are used to quantitatively evaluate the handwriting identification results of English handwriting.

[0076] 1) The formula for calculating the average accuracy rate:

[0077]

[0078] Among them, m represents the number of handwriting materials in the query database, n represents the handwriting materials to be queried, and R represents the total number of documents in the database that are the same as the i-th handwriting; P(k) represents the identification accuracy of the first k results, that is, the top k results The ratio of the number of related...

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 an English handwriting identification method based on improved VGG-16 model which comprises the following steps of collecting English document data sets from different handwriting of different persons; carrying out word segmentation on the obtained handwritten English handwriting document to obtain a data set composed of English words, and constructing a training sample setand a test sample set based on the data set; obtaining an improved VGG-16 model used for constructing a convolutional neural network, wherein the model comprises two traditional convolutional layers,three composite convolutional layers, five pooling layers and three full connection layers; inputting a training sample set, extracting character features, and carrying out classification training; and automatically identifying the English handwriting by using the trained neural network. The identification of English handwriting can be efficiently realized and identification accuracy rate can reach 100%. An algorithm has certain robustness on illumination change, simple geometric deformation and additional noise, and can be used in related fields of handwriting English handwriting identification. The algorithm can also be used for handwriting identification of other character handwritings after being expanded.

Description

technical field [0001] The invention relates to the fields of computer vision and pattern recognition, in particular to an English handwriting identification method based on the improved VGG-16 model, which can identify handwritten English and can be used to identify other handwritten handwritings after expansion. Background technique [0002] Handwriting identification is a test technique to determine whether the handwriting is the same by comparing the similarity between the text to be tested and the sample handwriting. It is widely used in many fields such as judicial identification, forensic science, and financial contract confirmation. The current note identification can be divided into two major directions: online handwriting identification and offline handwriting identification; among them, offline handwriting identification can be divided into two categories: based on local features and global features. The local feature-based method is to describe the local structur...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V40/33G06N3/045G06F18/24
Inventor 何凯马红悦冯旭刘坤
Owner TIANJIN UNIV
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