基於機器學習的手寫多語言與數學辨識系統

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：31

、訪客IP：18.218.84.153

姓名

古明翰(Ming-Han Ku) 查詢紙本館藏

畢業系所

資訊工程學系

論文名稱

基於機器學習的手寫多語言與數學辨識系統
(A Machine Learning Based System for Multi-Language and Mathematical Online Handwriting Recognition)

相關論文

★ 具多重樹狀結構之可靠性群播傳輸	★ 在嵌入式行動裝置上設計與開發跨平台Widget
★ 在 ARM 架構之嵌入式系統上實作輕量化的手持多媒體播放裝置圖形使用者介面函式庫	★ 基於網路行動裝置所設計可擴展的服務品質感知GStreamer模組
★ 針對行動網路裝置開發可擴展且跨平台之GSM/HSDPA引擎	★ 於單晶片多媒體裝置進行有效率之多格式解碼管理
★ IMS客戶端設計與即時通訊模組研發：個人資訊交換模組與即時訊息模組實作	★ 在可攜式多媒體裝置上實作人性化的嵌入式小螢幕網頁瀏覽器
★ 以IMS為基礎之及時語音影像通話引擎的實作:使用開放原始碼程式庫	★ 電子書嵌入式開發: 客制化下載服務實作, 資料儲存管理設計
★ 於數位機上盒實現有效率訊框參照處理與多媒體詮釋資料感知的播放器設計	★ 具數位安全性的電子書開發：有效率的更新模組與資料庫實作
★ 適用於異質無線寬頻系統的新世代IMS客戶端軟體研發	★ 在可攜式數位機上盒上設計並實作重配置的圖形使用者介面
★ Friendly GUI design and possibility support for E-book Reader based Android client	★ Effective GUI Design and Memory Usage Management for Android-based Services

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

至系統瀏覽論文 ( 永不開放)

摘要(中)

隨著時代的發展，交換信息的方式也在不斷的變化，人與人的交流越來
越方便。人們從最初的書信交流轉換為使用手機，平板電腦，計算機和其
他工具。網際網路不僅改變了人們之間的交流方式，也改變了傳統的教育
方式。不限時間地點的在線學習課程越來越受歡迎，許多數位學習平台也
大量出現。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紀錄知識。由於越來越多的用戶放棄了過往的紙筆作業，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。目前市面上也有許多成熟的手寫識別系統，但他們卻也有很大的侷限性。在單一語種的辨識具有較高的準確性，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，需要先進行文字區域偵測，並識別語言種類，找到用同一語言書寫的文本部分，再將其放入特定語言的識別系統。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，我們將致力於建立一個可以正確識別多種語言手寫文本的系統。

摘要(英)

With the development of the times, the way of exchanging information is constantly changing, and the communication between people is becoming more and more convenient. People have transformed from the initial communication of letters to the use of mobile phones, tablets, computers and other tools. The Internet has not only changed the way people communicate, but also the traditional way of education. Unlimited online learning courses are becoming
more and more popular, and many digital learning platforms are also appearing in large numbers. People began to use mobile phones and tablet computers for learning, and smart pens began to be widely used. Learners use smart pens to
take notes or record knowledge on electronic devices. As more and more users abandon their past paper-and-pen assignments, and instead use smart pens for writing. Therefore, the handwriting recognition technology has become more and more important. For multilingual countries and regions, the number of handwritten documents in multiple languages and scripts is gradually increasing.
There are many mature handwriting recognition systems on the market, but they also have great limitations. The recognition in a single language has high accuracy, but 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 a multi-language document, it is necessary to first detect the text area and identify the language type, find the text part written in the same language, and then put it into the recognition system of the specific language. In order to recognize more
and more handwritten documents in languages and scripts, and allow users to be unrestricted when writing documents. Therefore, in this project, we will strive to establish a system that can correctly recognize handwritten text in multiple languages.

關鍵字(中)

★ 手寫辨識
★ 機器學習
★ 圖像切割
★ 文本分類
★ 文字偵測

關鍵字(英)

★ Handwriting Recognition
★ Machine Learning
★ Image Segmentation
★ Script Classification
★ Text Detection

論文目次

摘要 II
Abstract III
致謝 V
Table of Contents VI
List of Figures VIII
List of Tables IX
1. Introduction 1
1.1 Background 1
1.2 Research Goal 2
2. Related Works 5
3. System Design 10
3.1 Pre-Processing 11
3.1.1 Smoothing Strokes 12
3.1.2 Document Segmentation 12
3.1.2.1 Text Area Detection 13
3.1.2.2 Finding character components of each line 15
3.1.2.3 Merge character components 16
3.1.3 Deleted text processing 19
3.2 Feature Extraction 20
3.2.1 Chinese Character 20
3.2.2 Mathematical Structure 21
3.3 Script Identification 22
3.3.1 Script Identification using Convolutional Neural Networks 22
3.3.2 KNN for proofreading 24
3.3.3 Context verification 24
3.3.4 Mathematical expression 25
3.4 Recognition 27
4. Evaluation 29
5. Conclusion and Future Work 32
Reference 34

參考文獻

[1] V. Carbune, P. Gonnet, T. Deselaers, H. A. Rowley,A. Daryin, M. Calvo, L.-L. Wang, D.
Keysers, S. Feuz, and P. Gervais, “Fast multi-language lstm-based online handwriting recognition,”
International Journal on Document Analysis and Recognition (IJDAR), pp. 1–14, 2020.
[2] K. Ubul, G. Tursun, A. Aysa, D. Impedovo, G. Pirlo, and T. Yibulayin, “Script
identification of multi-script documents: A survey, ”IEEE Access, vol. 5, pp. 6546–6559,
2017.
[3] N. Zheng, Q. You, G. Meng, J. Zhu, S. Du, and J. Liu,“50 years of image processing and
pattern recognition in china, ”IEEE Intelligent Systems, vol. 23, no. 6, pp.33–41, 2008.
[4] A. Graves, M. Liwic i, S. Fern ande , . ertolami, H. Bunke, and J. Schmidhuber, “A novel
connectionist system for unconstrained handwriting recognition,” IEEE Transactions on Pattern Analysis
and Machine Intelligence, vol. 31, no. 5, pp. 855–868, 2009.
[5] J. Puigcerver, “Are multidimensional recurrent layers really necessary for handwritten text recognition?”
in 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), vol. 1.
IEEE, 2017, pp. 67–72.
[6] D. Ghosh, T. Dube, and A. Shivaprasad, “Script recognition—a review,” IEEE Transactions on pattern
analysis and machine intelligence, vol. 32, no. 12, pp. 2142–2161, 2010.
[7] Z. Tian, W. Huang, T. He, P. He, and Y. Qiao, “Detecting text in natural image with connectionist text
proposal networ ,” in European conference on computer vision. Springer, 2016, pp. 56–72.
[8] X. Zhou, C. Yao, H. Wen, Y. Wang, S. Zhou, W. He, and J. Liang, “East: an efficient and accurate scene
text detector,” in Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, 2017,
pp. 5551–5560.
[9] . Shi, X. ai, and S. elongie, “Detecting oriented text in natural images by lin ing segments,” in
Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 2550–2558.
[10] M. Liao, B. Shi, X. Bai, X. Wang, and W. Liu, “Textboxes: A fast text detector with a single deep neural
networ ,” in Proceedings of the AAAI conference on artificial intelligence, vol. 31, no. 1, 2017.
[11] M. Liao, . Shi, and X. ai, “Textboxes++: A single-shot oriented scene text detector,” IEEE
transactions on image processing, vol. 27, no. 8, pp. 3676–3690, 2018.
[12] J. Ma, W. Shao, H. Ye, L. Wang, H. Wang, Y. Zheng, and X. Xue, “Arbitrary-oriented scene text detection
via rotation proposals,” IEEE Transactions on Multimedia ,vol. 20, no. 11, pp. 3111–3122, 2018.
[13] W. Wang, E. Xie, X. Li, W. Hou, T. Lu, G. Yu, and S. Shao, “Shape robust text detection with
progressive scale expansion networ ,” in Proceedings of the IEEE/CVF Conference on Computer Vision
and Pattern Recognition, 2019, pp. 9336–9345.
[14] Y. Jiang, X. Zhu, X. Wang, S. Yang, W. Li, H. Wang, P. Fu, and Z. Luo, “ 2cnn:
rotational region cnn for orientation robust scene text detection, ”arXiv preprintar Xiv : 1706.09579, 2017
[15] A. M. Namboodiri and A. K. Jain, “Online handwritten script recognition, ”IEEE Transactions on Pattern
Analysis and Machine Intelligence, vol. 26, no. 1, pp. 124–130,2004.
[16] D. Keysers, T. Deselaers, H. A. Rowley, L. Wang, and V. Carbune, “Multi-language online
handwriting recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no.
6, pp. 1180–1194, 2017.
[17] K. oy and K. Majumder, “Trilingual script separation of handwritten postal document,” in 2008 Sixth
Indian Conference on Computer Vision, Graphics Image Processing, 2008, pp. 693–700.
[18] N. Nikolaou, M. Makridis, B. Gatos, N. Stamatopoulos, and N. Papamar os, “Segmentation of historical
machine-printed documents using adaptive run length smoothing and s eleton segmentation paths,” Image
and Vision Computing, vol. 28, no. 4, pp. 590–604, 2010.
[19] M. Ma, D.-W. Park, S. K. Kim, and S. An, “Online recognition of handwritten korean and english
characters.” JIPS, vol. 8, no. 4, pp. 653–668, 2012.
[20] T. Makkar, Y. Kumar, A. K. Dubey, Rocha, and A. Goyal, “Analogi ing time complexity of nn and cnn
in recogni ing handwritten digits,” in 2017 Fourth International Conference on Image Information
Processing (ICIIP), 2017, pp. 1–6.
[21] T. Suzuki and A. Fujii, “Mathematical document categori ation with structure of mathematical
expressions,” in 2017 ACM/IEEE Joint Conference on Digital Libraries (JCDL). IEEE, 2017, pp. 1–10.
[22] U. Garain and . . Chaudhuri, “ ecognition of online handwritten mathematical expressions,” IEEE
Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), vol. 34, no. 6, pp. 2366–2376,
2004.
[23] Y.-C. Wu, F. Yin, and C.-L. Liu, “Improving handwritten chinese text recognition using neural network
language models and convolutional neural networ shape models,” Pattern Recognition, vol. 65, pp. 251–
264, 2017.
[24] X.-Y. Zhang, F. Yin, Y.-M. Zhang, C.-L. Liu, and Y. Ben-gio, “Drawing and recognizing chinese
characters with recurrent neural networ ,” IEEE transactions on pattern analysis and machine
intelligence, vol. 40, no. 4, pp. 849–862, 2017.
[25] Y.-C. Wu, F. Yin, Z. Chen, and C.-L. Liu, “Handwritten chinese text recognition using separable
multi-dimensional recurrent neural networ ,” in 2017 14th IAPR International Conference on Document
Analysis and Recognition (ICDAR), vol. 1. IEEE, 2017, pp. 79–84.
[26] K. Chen, L. Tian, H. Ding, M. Cai, L. Sun, S. Liang, and Q. Huo, “A compact cnn-dblstm based
character model for online handwritten chinese text recognition,” in 2017 14th IAPR International
Conference on Document Analysis and Recognition (ICDAR), vol. 1. IEEE, 2017, pp. 1068–1073.
[27] K. He, X. Zhang, S. en, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of
the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778.
[28] Po-Chuan Chen, “Traditional chinese handwriting dataset“
https://github.com/AI-FREE-Team/Traditional-Chinese-Handwriting-Dataset, 2020

指導教授

吳曉光(Eric Hsiao-Kuang Wu)

審核日期

2021-8-27

推文