臺灣博碩士論文加值系統

　　文字辨識技術一直是學術界研究的一個熱門主題，其應用的範疇非常之廣泛。在當今許多科技產品中，文字辨識也已經成為必備的功能之一。本論文結合了CMOS影像感測器、Altera DE2開發板以及LCD液晶顯示器，以FPGA純硬體數位電路的方式，設計出一套手寫數字影像辨識系統。
　　本系統先以色彩空間轉換找到代表手部特徵的膚色範圍，將二值化後的影像經過型態學運算濾除雜訊，再求出影像中指尖點座標，記錄與分析手寫時指尖點移動之相對位置與軌跡向量，並比對數字0~9之特徵值判別條件式後，最後再即時地輸出手寫數字影像辨識結果。
　　本系統的特色在於操作者不需觸碰到任何裝置，並採平行運算處理，有別於其他觸控式與離線式數字辨識系統。與其他以PC作業系統為平台的文字辨識系統相比較，本系統具有處理速度快、體積微小化以及節省資源的優點。經實驗測試結果手寫數字0~9的平均辨識成功率為96%。

Character recognition technology has been an active research topic, and its application is very wide and extensive. In many of new technology products, character recognition has been one of the essential functions. This thesis aims to utilize Altera DE2 FPGA development board with a CMOS image sensor and a LCD monitor, designing a vision-based fingertip digits writing recognition system on a single FPGA chip.
First, we transform the color image to a binary image by finding the range of skin color representing the hand characteristic, and then use morphological operations to filter out binary image noise. Second, the fingertip coordinate is recoding and analyzing the relative locations and locus vectors when fingertip moved. After that, the characteristic conditions of digits 0~9 are verified individually. Finally, the system output the result of recognition immediately.
The advantages of this system is that the operator does not need a touching device, and the system hardware is parallel processing, which are different from other touch panel and off-line digits recognition system. When comparing with the previous PC-based character recognition systems, its speed is faster, its size is smaller, and its hardware elements is less. The experimental result shows that the recognition successful rate for digits 0~9 is 96% in average.

摘要............................................I
Abstract.......................................II
致謝..........................................III
目錄...........................................IV
圖表索引.....................................VIII
第一章 緒論.....................................1
1.1 研究動機與目的.............................1
1.2 文獻回顧...................................2
1.3 系統描述與概觀.............................3
1.4 論文架構..................................10
第二章 手寫數字影像辨識系統相關理論基礎........11
2.1 色彩空間..................................11
2.1.1　RGB色彩空間...........................11
2.1.2　正規化RGB色彩空間.....................12
2.1.3 HSV色彩空間...........................12
2.1.4 YCbCr色彩空間.........................13
2.2 二值化....................................14
2.3 型態學....................................15
2.3.1 侵蝕與膨脹.............................15
2.3.2 平滑線性濾波...........................18
2.3.3 排序濾波...............................20
2.3.4 輪廓偵測...............................21
2.4 數字特徵辨識..............................23
2.4.1 統計式特徵.............................23
2.4.2 結構式特徵.............................25
第三章 即時手寫數字影像辨識系統................26
3.1 手部資訊分離..............................26
3.1.1　膚色偵測..............................26
3.1.2 型態學處理.............................32
3.2 指尖點偵測與定位..........................35
3.3 手寫軌跡重建..............................37
3.4 手寫數字辨識法............................37
3.4.1 數字特徵點數值化.......................38
3.4.2 數字軌跡向量化.........................40
3.5 基於決策樹理論之分類器設計................44
第四章 系統模組實現............................47
4.1 FPGA系統架構..............................47
4.2 影像處理功能模組..........................48
4.2.1 CMOS Sensor Controller模組.............48
4.2.2 影像前處理模組.........................50
4.2.3 型態學處理模組.........................57
4.3 SDRAM CONTROLLER模組......................65
4.4 指尖點偵測模組............................67
4.5 手寫數字辨識模組..........................68
4.5.1數字辨識啟動模組........................68
4.5.2 手寫軌跡重建模組.......................70
4.5.3 數字特徵比對與辨識模組.................70
4.6 VGA CONTROLLER模組........................78
第五章 實驗結果................................81
5.1 即時影像處理驗證..........................81
5.2 手寫數字辨識結果..........................85
5.3 FPGA資源分配..............................88
第六章 結論與建議..............................90
6.1 結論......................................90
6.2 建議......................................91
參考文獻.......................................93
作者簡介.......................................96

[1] 郭癸蘭，手寫身份證字號辨識系統，高雄第一科技大學電腦與通訊工程研究所碩士論文，民國九十一年。
[2] 黃偉詮，應用影像處理技術於統一發票之號碼自動辨識，國立台灣科技大學高分子工程研究所碩士論文，民國九十七年。
[3] 張簡子介，用小腦模型在FPGA上作車牌辨識，國立台灣師範大學工研教育研究所碩士論文，民國九十一年。
[4] L. Jin and D. Yang, "A novel vision based finger-writing character recognition system," Journal of Circuits, Systems, and Computers, Vol. 16, pp. 421-436, 2007.
[5] T. Long, L. W. Jin, L. X. Zhen and J. C. Huang, "One stroke cursive character recognition using combination of directional and positional features," IEEE International Conference on Acoustics, Speech, and Signal Processing, Vol. 5, pp. v/449-v/452, 2005.
[6] A. Sanghi, H. Arora1, K. Gupta1 and V. B. Vats, "A fingertip detection and tracking system as a virtual mouse, a signature input device and an application selector," Proceedings of the 7th International Caribbean Conference on Devices, pp. 1-4, 2008.
[7] M. Soriano, S. Huovinen, B. Martinkauppi, and M. Laaksonen, "Using the skin Locus to cope with changing illumination conditions in color-based face tracking," Proceedings of IEEE Nordic Signal Processing Symposium, pp. 383-386, 2000.
[8] CCIR, "Encoding Parameters of Digital Television for Studios,"　CCIR Recommendation 601-2, Int. Radio Consult. Committee,　Geneva, Switzerland, 1990.
[9] R. C. Gonzalez and R. E. Woods, "Digital Image Processing, 2nd edition", Pearson Education Inc., 2002.
[10] 鐘國亮，影像處理與電腦視覺第四版，東華書局，台北，2008。
[11] X. Xu, E. L. Miller, D. Chen and M. Sarhadi, "Adaptive two-pass rank order filter to remove impulse noise in highly corrupted images," IEEE Transactions on Image Processing, Vol. 13, No.2, pp. 238-247, 2004.
[12] O. D. Trier, A. K. Jain and T. Taxt, "Feature extraction methods for character recognition - a survey," Pattern Recognition, Vol. 29, No. 4, p. 641-662, 1996.
[13] Z. Chi and H. Yan, "ID3-derived fuzzy rules and optimized defuzzification for handwritten numeral recognition," IEEE Transactions on Fuzzy Systems, Vol. 4, Issue 1, pp. 24-31, 1996.
[14] R. Plamondon and S. N. Srihari, "On-line and off-line handwriting recognition: a comprehensive survey, " IEEE Transactions on Pattern Analysis And Machine Intelligence, Vol. 22, No.1, 2000.
[15] A. K. Jain, "Biometric authentication: some new research directions," Michigan State University, USA.
[16] M. H. Yang, D. J. Kriegman, and N. Ahuja, "Detecting faces in images: a survey," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, pp. 34 -58, 2002.
[17] Y. Zheng, G. Yang and L. Wu, "Fast face detection in field programmable gate array," IEEE International Conference on Digital Manufacturing and Automation, Vol. 1, pp. 719-723, 2010.
[18] P. S. Hiremath and A. Danti, "Detection of multiple faces in an image using skin color information and lines-of-separability face model," International Journal of Pattern Recognition and Artificial Intelligence, Vol. 20, No. 1, pp. 39-61, 2006.
[19] Y. Wand and B. Yuan, "A novel approach for human face detection from color image under complex background," Pattern Recognition, Vol. 34, No.10, pp. 1983-1992, 2001.
[20] D. Chai and A. Bouzerdoum, "A Bayesian Approach to Skin Color Classification in YCbCr Color Space," IEEE Region Ten Conference, Vol. 2, pp. 421-424, 2000.
[21] 友晶科技，DE2_User_manual。
[22] C. D. Ruberto and A. Morgera, "Moment-based techniques for image retrieval," IEEE 19th International Conference on Database and Expert Systems Application, 2008.
[23] M.B. Lin, "Digital System Design and Practices Using Verilog HDL and FPGAs," John Wiley & Son Inc., 2008.