基於手寫數學應用題內容利用卷積神經網路之輕量級多語種分類實作;Lightweight multilingual classification using convolutional neural network based on handwritten math application content Development

NCUIR > College of Electrical Engineering & Computer Science > Department of Computer Science and information Engineering > Research Project > Item 987654321/84702

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/84702

Title:	基於手寫數學應用題內容利用卷積神經網路之輕量級多語種分類實作;Lightweight multilingual classification using convolutional neural network based on handwritten math application content Development
Authors:	吳曉光
Contributors:	資訊工程學系
Keywords:	電腦視覺;手寫辨識;機器學習;智慧筆;圖像切割;多語言文本;Python;Computer Vision;Handwriting recognition;machine learning;Neo SmartPen;image segmentation;Multi-script;Python
Date:	2020-12-08
Issue Date:	2020-12-09 10:44:53 (UTC+8)
Publisher:	科技部
Abstract:	隨著時代的發展，人們交換信息的方式越來越方便。人們從最初的紙筆對話轉換為使用手機，平板電腦，計算機和其他工具。互聯網不僅改變了人們之間的交流方式，而且改變了傳統的教育方式。在線學習課程越來越受歡迎，許多數位學習平台正在逐步發展。人們開始使用手機和平板電腦進行學習，並且智能筆開始被廣泛使用，學習者利用智能筆在電子設備上做筆記或寫問題和答案。由於越來越多的用戶放棄了過去的紙筆作業，轉而使用智慧筆進行書寫。因此，手寫辨識這項技術變得越來越重要，對於多語言國家、地區，使用多種語言、文字的手寫文件數量也在逐步的增加。目前市場上也有許多成熟的手寫識別系統，它們的準確性在單一語種得情況下可以達到較高的辨識度。但對於多語種的文本的辨識，依然還有很大的改善空間。圖像中的多腳本識別是基於內容的圖像檢索和多語言系統開發的重要方向。在多語言文檔中，需要先進行語言識別，找到用同一語言書寫的部分文本，再將其應用於特定於語言的識別系統。為了識別越來越多的語言、腳本的手寫文檔，並允許用戶在書寫文件時不受限制。因此本計畫中，我們將致力於正確識別這些多種語言的手寫文本。 ;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.
Relation:	財團法人國家實驗研究院科技政策研究與資訊中心
Appears in Collections:	[Department of Computer Science and information Engineering] Research Project

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	155	View/Open

社群 sharing

Loading...