Lightweight multilingual classification using convolutional neural network based on handwritten math application content Development

Project Details

Description

With the development of the times, the way people exchange information becomes more and more convenient. People switched from the initial paper-and-pencil conversation to using mobile phones, tablets, computers and other tools. The Internet has not only changed the way people communicate, but also changed the traditional way of education. Online learning courses are becoming more and more popular, and many digital learning platforms are gradually developing. People began to use mobile phones and tablets for learning, and smart pens began to be widely used. Learners used smart pens to take notes or write questions and answers on electronic devices. As more and more users give up past paper-and-pencil work, they use smart pens for writing. Therefore, the technology of handwriting recognition is becoming more and more important. For multilingual countries and regions, the number of handwritten files using multiple languages and characters is also gradually increasing.At present, there are many mature handwriting recognition systems on the market, and their accuracy can achieve a high degree of recognition in a single language. However, there is still much room for improvement in the recognition of multilingual texts. Multi-script recognition in images is an important direction for content-based image retrieval and multi-language system development. In multilingual documents, you need to perform language recognition first, find part of the text written in the same language, and then apply it to a language-specific recognition system. In order to recognize more and more handwritten documents in languages and scripts, and allow users to write files without restrictions. Therefore, in this project, we will work to correctly recognize these handwritten texts in multiple languages.
StatusFinished
Effective start/end date1/11/2031/10/21

UN Sustainable Development Goals

In 2015, UN member states agreed to 17 global Sustainable Development Goals (SDGs) to end poverty, protect the planet and ensure prosperity for all. This project contributes towards the following SDG(s):

  • SDG 4 - Quality Education
  • SDG 17 - Partnerships for the Goals

Keywords

  • Computer Vision
  • Handwriting recognition
  • machine learning
  • Neo SmartPen
  • image segmentation
  • Multi-script
  • Python

Fingerprint

Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.