研究生: |
徐銘伽 Ming-Cie Syu |
---|---|
論文名稱: |
以FPGA架構加速字元辨識演算法(以SOPC實現並驗證) FPGA Architecture for Accelerating Character Recognition Algorithm (Implementation and Verification on SOPC) |
指導教授: |
許孟超
Mon-Chau Shie |
口試委員: |
陳伯奇
none 梁文耀 none 阮聖彰 none |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 電子工程系 Department of Electronic and Computer Engineering |
論文出版年: | 2006 |
畢業學年度: | 94 |
語文別: | 中文 |
論文頁數: | 73 |
中文關鍵詞: | 字元辨識 |
外文關鍵詞: | Character Recognition, SOPC |
相關次數: | 點閱:111 下載:1 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
車牌辨識系統廣泛應用於交通流量的監督和監看,像是贓車協尋、停車場的控管和交通流量資訊等方面。一個車牌辨識系統,通常分成三大部分,一是車牌定位,車牌分割和字元辨識。本研究範圍為字元辨識部份,目前的字元辨識方法有類神經網路法[23-24, 30, 37]、結構法[26, 28, 41-42]、樣板比對法[6, 22, 32]還有一種是定點取樣法[25]。
本研究的字元辨識方法採用定點取樣法,此法是在圖像字元中幾個特別關鍵的區域,將這些關鍵區域用0(無邊經過)與1(有邊經過)的組合(特徵值)所表示出來的方法就稱為定點取樣法。研究者經過仔細觀察字(英文和數字)的特徵之後共觀察出了字的17個關鍵區域,而此17個關鍵區域中再經過本研究的判定(0或1)法則後產生出一筆17個0或1所組成的數字(稱為特徵值), 再將此特徵值與特徵表(資料庫)比對,最後比對出該車牌字元的號碼。此字元辨識演算法簡單容易實做成硬體且也比其它方法較節省硬體資源,其辨識率也不差於一般的字元辨識演算法。
本研究的字元辨識以SOPC(System on Programmable Chip)為平台,來逹到輕巧、低成本的可攜式的車牌子元辨識設備。SOPC( System On Programmable Chip)為一個FPGA,內含有一個32-bit RISC處理器,可以輕易的加入自行定義的硬體模組與處理器結合。SOPC具有完善的硬體模組函式庫,良好撰寫C程式語言的使用者介面,具有硬體執行速度快的特性與軟體容易修改的特點,在開發上比ASIC具彈性,在消耗功率上比個人電腦低。
在字元辨識的前置處理(不包括車牌定位)部份雖然不是本研究之範圍,但為了呈現一個完整的車牌辨識系統還是用SOPC軟體來實現此部份功能。本論文最後針對60張車牌即360張圖像字元中作測試,經實驗證實車牌辨識率90%(成功辨識54個),字元辨識率97%(成功辨識352個)。在速度方面研究者使用SOPC提供FPGA將整個字元辨識做成硬體(除了有些的相似字元比較部份為軟體),經研究証明改為硬體後速度約提高5.1倍。在硬體資源方面,其整個字元辨識硬體僅需875個邏輯單元(Logic cells)和使用2.176 kbits的記憶體,且與其它的論文[22-23, 37]的字元辨識硬體比較後証明本研究的字元辨識硬體具有超節省硬體資源。
License plate recognition system is widely used in monitoring information of traffic, such as searching stolen car, parking lot control, and information on traffic flow. A license plate recognition system usually separated into three parts, that is, license plate position, license plate division, and character recognition. This research topic focuses on character recognition. So far, the character recognition processing has some technique to implement, such as neural networks [23-24, 30, 37], syntactic structural [26, 28, 41-42], template matching [6, 22, 32] and fixed-point sampling method[25].
This paper proposes a new character recognition algorithm, which employs the fixed-point sampling method.We discovers other critical regions and proposed new decision logic to these critical regions. According to the experiments, seventeen critical regions are discovered, and the proposed decision logic is applied to these regions to generate characteristic values. Matching characteristic values with database recognizes the meaning number of the character. Also, the algorithm which has good recognition rate is implemented in hardware. Compare with other approaches, this proposed architecture do not need too many hardware resource.
SOPC combines a 32-bit RISC microprocessor and a FPGA. It’s flexible in development and low consumption in power. This paper presents an implementation of a lightweight, low cost, and portable license plate character recognition system based on SOPC platform.
Although the proposed algorithm doesn’t include the pre-processing of character recognition, we implemented this part by using software implementation. According to experiments, the proposed system achieves 90% license plate recognition rate, 97% character recognition rate, and hardware implemented character recognition delivers 5.1 times boost of recognition speed. Also the proposed hardware implementation requires 875 logic cells and 2.176 kbits memory only, which is very efficient of saving hardware resource in comparison of other implementations[22-23, 37].
[1] N. Otsu, “A threshold selection method from Gray-Level Histograms,” IEEE Trans. on System, Man and Cybernetics, vol. SMC-9, no. 1, pp. 62-66, 1979.
[2] Donald Hearn, M. Pauline Baker, Computer Graphics, Prentice-Hall International, London, 1986.
[3] B.K. Jang, R.T. Chin, “Analysis of thinning algorithms using mathematical morphology,” IEEE Transactions on Volume 12, Issue 6, pp. 541-551, June 1990.
[4] C.A. Rahman, W. Badawy and A. Radmanesh, “A real time vehicle's license plate recognition system,” IEEE Conference on Advanced Video and Signal Based Surveillance, pp. 163-166, July 2003.
[5] Y.T. Hsu, C.B. Lin, S.C. Mar, and S.F. Su, “High noise vehicle plate recognition using gray system,” Journal of Grey Systems, vol. 10, No. 3, pp. 193-208, 1998.
[6] P. Comelli, P. Ferragina, M.N. Granieri, and F. Stabile, “Optical recignition of motor vehicle license plates,”IEEE Trans. on Vehicular Tech., vol. 44, No. 4, 1995.
[7] S. Mori, C.Y. Suen, and K. Yamamoto, “History review of OCR research and development,”IEEE Proceedings, vol. 80, No. 7, pp. 1029-1058, 1992.
[8] S.K. Kim, D.W. Kim, and H.J. Kim, “A recognition of vehicle license plate using a genetic algorithm based segmentation,”Proceedings, International Conference on Image Processing, vol. 2, pp. 661-664, 1996.
[9] J.C.H. Poon, M. Ghadiali, G.M. T. Mao, and L.M. Sheung, “A robust vision system for vehicle licence plate recognition using gray-scale morphology,” Proceedings on the IEEE international Symposium on Industrial Electronics (ISIE’95), vol. 1, pp. 394-399, 1995.
[10] S.L. Chang, L.S. Chen, Y.C. Chung, and S.W. Chen, “Automatic license plate recognition,” IEEE Transactions on Intelligent Transportation Systems, vol. 5, pp. 42-53, Mar. 2004.
[11] N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Transactions on System, vol. SMC-9, pp. 62-66, 1979.
[12] S. Mori, C.Y. Suen, K. Yamamoto, “Historical review of OCR research and development,” Proceedings of the IEEE Volume 80, issue 7, pp. 1029-1058, July 1992.
[13] R.G. Casey, E. Lecolinet, “A survey of methods and strategies in character segmentation,” IEEE Transactions on Volume 18, issue 7, pp. 690-706, July 1996.
[14] T. Chang, C.C.J. Kuo, “Texture segmentation with tree-structured wavelet transform,” Time-Frequency and Time-Scale Analysis, pp. 543-546, Oct. 1992.
[15] T. Chang, C.C.J. Kuo, “Texture analysis and classification with tree-structured wavelet transform,” Image Processing, pp. 429-441, Oct. 1993.
[16] Y.T. Yuan, F.L. Bing, M. Hong, L. Jiming, C.H. Leung, Ching Y. Suen, “A novel approach to optical character recognition based on ring-projection-wavelet-fractal signatures,” Pattern Recognition, vol. 2, pp. 325-329, Aug. 1996.
[17] V. DeBrunner, M. Kadiyala, “Effect of wavelet bases in texture classification using a tree-structured wavelet transform,” Signals, Systems, and Computers, vol. 2, pp. 1292-1296, Oct. 1999.
[18] T.M. Ha, “Efficient detection of abnormalities in large OCR databases,” Document Analysis and Recognition, vol. 2, pp. 1006-1010, Aug. 1997.
[19] S. Tsujimoto, H. Asada, “Major components of a complete text reading system,” Proceedings of the IEEE., vol. 80, issue 7, pp. 1133-1149, July 1992.
[20] J. Gllavata, R. Ewerth, B. Freisleben, “A text detection, localization and segmentation system for OCR in images,” Multimedia Software Engineering, IEEE Sixth International Symposium on 13-15, pp. 310-17, Dec. 2004.
[21] M. Tellache, M. Sid-Ahmed, B. Abaza, “Thinning algorithms for Arabic OCR,” Communications, Computers and Signal Processing, IEEE Pacific Rim Conference on Volume 1, pp. 248-251, May. 1993.
[22] A. Ahmadi, Y. Shirakawa, M.A. Abedin, K. Kamimura, H.J. Mattausch, and T. Koide, “An LSI hardware design for online character recognition using associative memory,” IEEE Cnf. on Circuits and Systems, vol. 1, pp. 464-467, Aug. 2005.
[23] L. Carro, and D.T. Franco, “FPGA based systems with linear and non-linear signal processing capabilities,” Euromicro Conference, vol. 1, pp. 260-264, Sept. 2000.
[24] E. Sackinger, B. Boser, J. Bromley, Y. LeCun, and L.D. Jackel, “Application of the
ANNA neural network chip to high-speed character recognition,” IEEE Trans. Neural Networks, vol. 3, no. 3, pp. 498-505, May 1992.
[25] H. Selvaraj, M. Venkatesan, “A reconfigurable printed character recognition system
using a logic synthesis tool,” Euromicro Conference, Volume 1, pp. 24-29, Aug. 1998.
[26] L. Jianzhuang, L. Wenqing, and T. Yupeng, “Automatic threshoding of gray-level pictures using two-dimension otsu method,” International Conf. on Circuits and Systems, vol. 1, pp. 325-327, June 1991.
[27] T. Naito, T. Tsukada, K. Yamada, K. Kozuka, and S. Yamamoto, “Robust License-Plate Recognition Method for Passing Vehicles under Outside Environment,” IEEE Trans. on Vehicular Technology, vol. 49, no. 6, Nov. 2000.
[28] P. Comelli, P. Ferragina, M.N. Granieri, and F. Stabile, “Optical recognition of motor vehicle license plates,” IEEE Trans. on Vehicular Technology, vol. 44, no. 4, Nov. 1995.
[29] H.H. Lin, C.Y. Chen, and J.H. Chuang, “Recognition of printed digits of low resolution,” Pattern Recognition and Image Analysis, vol. 10, no. 2, pp. 265-272, Dec. 2000.
[30] R. Parisi, E.D. Di Claudio, G. Lucarelli, and G. Orlandi, “Car plate recognition by neural networks and image processing,” Proceedings of the 1998 IEEE International Symposium on Circuits and Systems, vol. 3, pp. 195-198, 1998.
[31] N. Mani, and B. Srinivasan, “Application of artificial neural network model for optical character recognition,” IEEE International Conference on Systems, vol. 3, pp. 2517-2520, 1997.
[32] H.A. Hegt, R.J. Haye, N.A. Khan, “A high performance license plate recognition system,” IEEE Conference on Systems, vol. 5, pp. 4357-4362, 1998.
[33] Y.P. Huang, S.Y. Lai, and W.P. Chuang, “A template-based model for license plate recognition,” IEEE International Conference on Networking, vol. 2, pp. 737-742, March 2004.
[34] R. Bozinovic, and S.N. Srihari, “String Correction Algorithm for Cursive Script Recognition,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 4, no. 6, pp. 655-663, June 1982.
[35] H. Fujisawa, Y. Nakano, and K. Kurino, “Segmentation Methods for Character Recognition: From Segmentation to Document Structure Analysis,” Proc. IEEE, vol. 80, no. 7, pp. 1,079-1,092, July 1992.
[36] 林仲芬,影像辨認技術,二版,全華科技,台北,民國八十四年。
[37] 張簡子,「以小腦模型在FPGA 上作車牌辨識」,國立台灣師範大學工業教育所,碩士論文,民國七十九年。
[38] 林欣平,“車牌字元粹取,”國立交通大學,電機與控制工程研究所碩士論
文,1999。
[39] 陳同孝,張真誠,黃國峰,“數位影像處理技術,”松崗電腦圖書,2000。
[40] 連國珍,“數位影像處理,”儒林圖書,1996。
[41] 周俊男,輛牌照影像辨識系統,國立中山大學資訊工程研究所,1995。
[42] 張銘豪,用分割辨識方法之英文數字辨識系統,立中山大學資訊工程
研究所,1996。