基於Bi-LSTM的筆順辨識與筆跡標準度評估的數位學習系統;Bi-LSTM Based Recognition of Stroke Order and Handwriting Quality for E-Learning

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/93196

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/93196

Title:	基於Bi-LSTM的筆順辨識與筆跡標準度評估的數位學習系統;Bi-LSTM Based Recognition of Stroke Order and Handwriting Quality for E-Learning
Authors:	李政勳;Lee, Cheng-Hsun
Contributors:	資訊工程學系
Keywords:	雙向長短記憶網路;數位學習;手寫;筆跡標準度計算;筆順辨識;Bi-LSTM;e-learning;handwriting;handwriting quality calculation;stroke order recognition
Date:	2023-07-24
Issue Date:	2024-09-19 16:47:18 (UTC+8)
Publisher:	國立中央大學
Abstract:	得益於現今科技的快速發展以及互聯網的普及，數位學習如今已經成為許多人的學習媒介之一，有別於傳統教育受限於時間、地點以及教師資源，數位學習更可以提升學習效率。我們注意到這些多元的學習平台下，專門針對小學生的書寫問題進行自動辨識以及評估的平台並沒有很多的研究，且都仍有改善的空間，像是可能無法自動辨識筆順以及判斷筆跡的標準度，或是可能需要額外蒐集筆跡的裝置，小學生的線上學習手寫環境、資源目前是相對不友善的。對於小學生來說，學習書寫對於他們日後進行文本產生的相關複雜認知行為會產生重大的影響，因此學習好良好的書寫習慣對於小學生日後的發展是非常重要的。因此在本研究中，提出了利用雙向長短記憶網路來自動辨識小學生書寫英文字母筆順正確以及筆跡標準度的數位學習系統，透過使用智慧手寫筆寫於專用的紙上回傳資料至伺服器中進行評估並回傳結果。在實驗結果中，我們在辨識筆順是否正確上取得了非常好的準確度，此外在判斷筆跡標準度的時候我們系統所設計的評分標準也能夠有效的讓學生自我檢視學習成果，達到數位學習的提升效率的功能。;Due to the rapid development of technology and the popularity of the Internet, e-learning becomes one of the learning mediums for many people today. Compare to traditional education, which is limited by time, location, and teacher resources, e-learning can enhance learning efficiency. We noticed that there has not been much research on these multiple learning platforms that specifically recognize and evaluate primary school students′ handwriting problems, indicating potential for improvement, such as systems that may not automatically recognize stroke order and evaluate the handwriting quality, or systems that may require additional devices to collect handwriting. Currently, the e-learning environment and resources for primary school students are relatively unfriendly. For primary school students, learning to write has a significant impact on the complex cognitive behaviors associated with text production in the future, so learning good handwriting habits is very important for the development of primary school students. Therefore, in this study, we proposed an e-learning system that uses a bidirectional long short-term memory to automatically recognize the correct stroke order and handwriting quality of primary school students′ written English letters. By using a smartpen to write on a special paper, the data is transmitted to the server for evaluation and return of results. In the experimental results, we achieved outstanding accuracy in recognizing the correctness of stroke order. Additionally, our system′s designed scoring criteria for handwriting quality effectively enables students to self-assess their learning results, thus enhancing the efficiency of e-learning.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	27	View/Open

社群 sharing

Loading...