請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/68716完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 黃乾綱(Chien-Kang Huang) | |
| dc.contributor.author | Yu-An Chen | en |
| dc.contributor.author | 陳昱安 | zh_TW |
| dc.date.accessioned | 2021-06-17T02:31:57Z | - |
| dc.date.available | 2020-08-24 | |
| dc.date.copyright | 2020-08-24 | |
| dc.date.issued | 2020 | |
| dc.date.submitted | 2020-08-18 | |
| dc.identifier.citation | 1. Mori, S., C.Y. Suen, and K.J.P.o.t.I. Yamamoto, Historical review of OCR research and development. 1992. 80(7): p. 1029-1058. 2. Fowler, G.C. and T.D. Hughey, Reading machine. 1979, Google Patents. 3. Hu, X., et al. A printed Chinese character recognition method. in 2011 International Conference on Computer Science and Service System (CSSS). 2011. IEEE. 4. Kim, I.-J., J.-H.J.I.T.o.P.A. Kim, and M. Intelligence, Statistical character structure modeling and its application to handwritten Chinese character recognition. 2003. 25(11): p. 1422-1436. 5. Casey, R. and G.J.I.T.o.E.C. Nagy, Recognition of printed Chinese characters. 1966(1): p. 91-101. 6. Wang, N. Printed Chinese character recognition based on pixel distribution probability of character image. in 2008 International Conference on Intelligent Information Hiding and Multimedia Signal Processing. 2008. IEEE. 7. Kimura, F., et al., Modified quadratic discriminant functions and the application to Chinese character recognition. 1987(1): p. 149-153. 8. Leung, C., et al., A knowledge-based stroke-matching method for Chinese character recognition. 1987. 17(6): p. 993-1003. 9. Chan, K.-P., Y.J.I.t.o.s. Cheung, man,, and cybernetics, Fuzzy-attribute graph with application to Chinese character recognition. 1992. 22(1): p. 153-160. 10. Bai, Z.-L. and Q. Huo. A study on the use of 8-directional features for online handwritten Chinese character recognition. in Eighth International Conference on Document Analysis and Recognition (ICDAR'05). 2005. IEEE. 11. Zhang, J., et al., A method of neighbor classes based SVM classification for optical printed Chinese character recognition. 2013. 8(3): p. e57928. 12. Chi, X., et al. A simple method for Chinese license plate recognition based on support vector machine. in 2006 International Conference on Communications, Circuits and Systems. 2006. IEEE. 13. Khawaja, A., et al. Recognition of printed Chinese characters by using Neural Network. in 2006 IEEE International Multitopic Conference. 2006. IEEE. 14. Hu, X., et al. A printed Chinese character recognition method. in Computer Science and Service System (CSSS), 2011 International Conference on. 2011. IEEE. 15. Zhong, Z., L. Jin, and Z. Feng. Multi-font printed Chinese character recognition using multi-pooling convolutional neural network. in 2015 13th International Conference on Document Analysis and Recognition (ICDAR). 2015. IEEE. 16. Zhong, Z., L. Jin, and Z. Xie. High performance offline handwritten chinese character recognition using googlenet and directional feature maps. in 2015 13th International Conference on Document Analysis and Recognition (ICDAR). 2015. IEEE. 17. Zhong, Z., et al. Handwritten Chinese character recognition with spatial transformer and deep residual networks. in Pattern Recognition (ICPR), 2016 23rd International Conference on. 2016. IEEE. 18. Wang, T.-Q. and C.-L. Liu. Fully convolutional network based skeletonization for handwritten chinese characters. in Thirty-Second AAAI Conference on Artificial Intelligence. 2018. 19. Cireşan, D. and U. Meier. Multi-column deep neural networks for offline handwritten Chinese character classification. in Neural Networks (IJCNN), 2015 International Joint Conference on. 2015. IEEE. 20. Zhang, Y.J.C.S.D., Stanford University, Deep convolutional network for handwritten Chinese character recognition. 2015. 21. Shijie, J., et al. Research on data augmentation for image classification based on convolution neural networks. in 2017 Chinese automation congress (CAC). 2017. IEEE. 22. Song, X., et al. A handwritten Chinese characters recognition method based on sample set expansion and CNN. in 2016 3rd International Conference on Systems and Informatics (ICSAI). 2016. IEEE. 23. LeCun, Y., et al., Backpropagation applied to handwritten zip code recognition. 1989. 1(4): p. 541-551. 24. Krizhevsky, A., I. Sutskever, and G.E. Hinton. Imagenet classification with deep convolutional neural networks. in Advances in neural information processing systems. 2012. 25. He, K., et al. Deep residual learning for image recognition. in Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. 26. Wang, T.-Q., F. Yin, and C.-L. Liu. Radical-based Chinese character recognition via multi-labeled learning of deep residual networks. in 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR). 2017. IEEE. 27. Liu, C.-L., et al. CASIA online and offline Chinese handwriting databases. in 2011 International Conference on Document Analysis and Recognition. 2011. IEEE. 28. Song, X., et al. A handwritten Chinese characters recognition method based on sample set expansion and CNN. in Systems and Informatics (ICSAI), 2016 3rd International Conference on. 2016. IEEE. 29. Suryani, D., P. Doetsch, and H. Ney. On the benefits of convolutional neural network combinations in offline handwriting recognition. in Frontiers in Handwriting Recognition (ICFHR), 2016 15th International Conference on. 2016. IEEE. 30. Rosenfeld, A. and J.L.J.P.r. Pfaltz, Distance functions on digital pictures. 1968. 1(1): p. 33-61. 31. Mikołajczyk, A. and M. Grochowski. Data augmentation for improving deep learning in image classification problem. in 2018 international interdisciplinary PhD workshop (IIPhDW). 2018. IEEE. 32. Zhang, R., et al. ICDAR 2019 robust reading challenge on reading chinese text on signboard. in 2019 International Conference on Document Analysis and Recognition (ICDAR). 2019. IEEE. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/68716 | - |
| dc.description.abstract | 近年來，隨著人工智慧的快速發展，深度學習（Deep Learning）的技術也隨之蓬勃發展，並廣泛應用在各個領域，包括中文字元影像辨識（Chinese Character Recognition）。 本研究的目的在改善中文漢字之辨識模型建立問題，利用現有電腦系統內建的字型資源來產生文字影像，再經由一系列的影像處理來模擬真實環境中的影像型態，並調整影像內文字本體部分數值，使得在使用機器學習中的卷積神經網路（Convolutional Neural Networks）之技術時能更有效學習到文字架構特徵而非邊界像素點分布之特徵。 經由實驗結果顯示，使用本研究方法在現代報紙與民初晶報等印刷文件之辨識準確率分別為97.66%與78.21%，在CASIA 公開中文手寫測試集內達到63.15%之辨識準確率，以及在針對ICDAR-2019年ReCTS (Robust Reading Challenge on Reading Chinese Text on Signboard)競賽內之測試資料集，在使用官方提供之訓練資料額外加入本研究方法所產生之文字影像一同訓練，達到91.26%的辨識準確率，上述所提及之辨識表現優於現有OCR系統及方法。 | zh_TW |
| dc.description.abstract | In recent years, with the rapid development of artificial intelligence, deep learning technology has also been widely applied to various fields, including Chinese Character Recognition. The main purpose of this paper is to solve the problem of Chinese character recognition model building. By using the existing Chinese font resources in computer system to generate text images, and then use a series of image processing to simulate the image in the real environment and adjust the pixel value of text in image. That makes it more effective to learn the features of the text structure rather than the characteristics of the boundary pixel distribution when using the technology of Convolutional Neural Networks in machine learning We conduct our experiments with newspaper and the Jing Newspaper, the CASIA handwritten Chinese character public test set and the Chinese character of ICDAR-2019 ReCTS race testing dataset. The results show that the model and method we proposed in this paper can reach the accuracy of 97.66% on newspaper, 78.21% on Jing Newspaper, 63.15% on handwritten, and 91.26% on ICDAR ReCTS. Compared with the existing common OCR recognition software, our method can improve the accuracy. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-17T02:31:57Z (GMT). No. of bitstreams: 1 U0001-1708202009474100.pdf: 6690842 bytes, checksum: 5185efa56697f90bb4878cddb63c55dd (MD5) Previous issue date: 2020 | en |
| dc.description.tableofcontents | 中文摘要 i 英文摘要 ii 目錄 iii 圖目錄 iv 表目錄 v 第一章 緒論 1 1.1 研究背景與動機 1 1.2 研究目標 2 1.3 研究貢獻 4 1.4 論文架構 5 第二章 文獻探討與背景知識 6 2.1 傳統非機器學習中文字辨識 6 2.1.1 區塊像素機率分布 6 2.1.2 筆畫向量特徵 7 2.2 現代機器學習中文字辨識 8 2.2.1 支援向量機 8 2.2.2 類神經網路 9 2.3 影像資料增強 12 2.4 卷積神經網路 13 第三章 研究方法 15 3.1 問題定義 15 3.1.1 文件文字辨識問題 15 3.1.2 手寫文字辨識問題 16 3.1.3 環境文字辨識問題 18 3.2 中文字編碼 19 3.3 中文字選取範圍 20 3.4 電腦字體篩選 21 3.5 影像灰階 22 3.5.1 三原色光模式 22 3.5.2 灰階轉換設定 23 3.6 電腦系統生成多樣性影像 25 3.6.1 影像尺寸大小 25 3.6.2 影像數值調整 26 3.6.3 旋轉 29 3.6.4 仿射轉換 30 3.7 卷積神經網絡模型與系統架構 31 第四章 實驗結果與討論 32 4.1 實驗資料蒐集與建置 32 4.1.1 印刷文件測試集 32 4.1.2 手寫文件測試集 33 4.1.3 環境文字測試集 34 4.2 漸層調整數值方式對辨識的影響 35 4.2.1 字體影響 35 4.2.2 旋轉角度影響 37 4.2.3 斜切角度影響 38 4.3 測試集辨識結果與討論 39 4.3.1 文件文字辨識 39 4.3.2 手寫文字辨識 40 4.3.3 環境文字辨識 41 4.4 減少收集真實環境影像 42 4.4.1 加入 ICDAR 訓練資料 42 4.4.2 加入篩選後訓練資料 43 4.5 真實場景內罕見字辨識 44 第五章 結論與未來展望 46 5.1 結論 46 5.2 未來展望 47 參考文獻 48 | |
| dc.language.iso | zh-TW | |
| dc.title | 應用字型生成資料開發環境中文字辨識系統 | zh_TW |
| dc.title | Chinese character recognition system in life developed by applying font generation data | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 108-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 傅楸善(Chiou-Shann Fuh),張恆華(Herng-Hua Chang),林政宏(Cheng-Hung Lin) | |
| dc.subject.keyword | 中文字元影像辨識,影像處理,資料增強,深度學習,卷積神經網絡, | zh_TW |
| dc.subject.keyword | CCR,Image Processing,Data Augmentation,Deep Learning,Convolutional Neural Networks, | en |
| dc.relation.page | 50 | |
| dc.identifier.doi | 10.6342/NTU202003685 | |
| dc.rights.note | 有償授權 | |
| dc.date.accepted | 2020-08-19 | |
| dc.contributor.author-college | 工學院 | zh_TW |
| dc.contributor.author-dept | 工程科學及海洋工程學研究所 | zh_TW |
| 顯示於系所單位： | 工程科學及海洋工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| U0001-1708202009474100.pdf 目前未授權公開取用 | 6.53 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。