臺灣博碩士論文加值系統

傳統上，架設車牌辨識系統時，需花費大量成本於設備、系統整合以及後續維護等問題上，相較之下，嵌入式車牌辨識系統除了具有較低的成本外，還有易於維護更換等優勢。本文採用ADSP-BF561發展板開發嵌入式即時車牌辨識系統。我們的系統可分為三大子系統，分別為車牌定位、車牌字元切割與字元辨識等。車牌定位部份，車牌影像有頻繁的紋理變化與特別的外觀形狀，因此，我們可針對這些特性實踐車牌定位。在開始字元切割前，我們會先搜尋車牌中已存在之字元邊界，並以此類邊界為參考依據，估測其他不容易切割或相連之字元影像。字元辨識部份可分為兩階段，第一階段以標準字元影像所建立之輪廓與紋理等資料庫與實際車牌之輪廓與紋理互相比對，從中尋找最相似之字元影像，而第二階段則針對幾種容易混淆之字型再次做確認，並將其結果作為最終辨識結果。
最後，實驗結果顯示，我們的嵌入式即時車牌辨識系統車牌定位成功率可達97％，且系統能在10ms內完成單張影像的車牌定位與辨識，故可即時處理。

Traditionally, when building system for license plate recognition, need to solve problems, such as system integration and maintenance...etc, but the advantage of embedded system of license plate recognition to not only low cost but also easy maintenance. In this paper, ADSP-BF561 as a development platform for embedded real-time system for license plate recognition. Real-time system for license plate recognition can be divided into three subsystems, namely, license plate location, character segmentation and character recognition. Part of license plate location, we can check the texture and shape to determine whether the license plate. Part of character segmentation, we first determine the character boundary exists in license plate, and to such boundaries as reference, other characters in the boundary estimation. Part of character recognition can be divided into two stages, the first stage, established a standard character image contour and texture of the database and the actual license plate of the contour and texture compared with each other, find the smallest error as the recognition results, The second stage, again recognizes to easily confuse of character.
Finally, the experiment shows as a result, our license plate location success rate is 97%, and the system can complete the license plate location and recognition of an image in the 10 ms and satisfy real-time processing.

中文摘要 ...................................................................i
英文摘要 ..................................................................ii
誌謝 .................................................................iii
目錄 ..................................................................iv
表目錄 ..................................................................vi
圖目錄 .................................................................vii
第一章 緒論...............................................................1
1.1 研究背景...........................................................1
1.2 相關應用研究.......................................................1
1.3 研究動機與目的.....................................................3
1.4 論文架構...........................................................3
第二章 發展板系統架構.....................................................4
2.1 發展板平台簡介.....................................................4
2.2 硬體系統架構.......................................................5
2.2.1 核心架構...........................................................5
2.2.2 Memory.............................................................7
2.2.3 DMA................................................................8
2.2.4 Interrupt.........................................................10
2.2.5 ITU-656...........................................................11
2.2.6 PPI...............................................................14
2.3 記憶體的運用與規劃................................................16
第三章 即時車牌辨識系統之架構及原理......................................20
3.1 即時車牌辨識系統架構..............................................20
3.2 車牌定位..........................................................22
3.3 車牌切邊..........................................................27
3.4 車牌字元切割......................................................33
3.5 車牌字元辨識......................................................37
第四章 實驗結果..........................................................41
4.1 車牌定位實驗......................................................41
4.2 車牌切邊實驗......................................................43
4.3 字元辨識實驗......................................................43
第五章 結論與未來展望....................................................47
5.1 結論..............................................................47
5.2 未來展望..........................................................47
參考文獻 ..................................................................48

[1] Analog Devices, ADSP-BF561 Blackfin? Processor Hardware Reference, Rev.
1.1, Analog Devices, USA, 2007.
[2] Lifeng He, Yuyan Chao, and Kenji Suzuki, “A Run-Based Two-Scan Labeling
Algorithm,” IEEE Transactions on Image Processing, vol. 17, no. 5, pp.
749-756, May, 2008.
[3] Lifeng He, Yuyan Chao, Kenji Suzuki, and Kesheng Wu , “Fast connected-
component labeling,” Pattern Recognition, vol. 42, pp. 1977-1987,
September, 2009.
[4] Kaushal Sanghai, Video Templates for Developing Multimedia Applications on
Blackfin? Processors, Analog Devices, September, 2006.
[5] Kaushal Sanghai, System Optimization Techniques for Blackfin? Processors,
Analog Devices, July, 2007.
[6] Kunal Singh, and Ramesh Babu, Video Framework Considerations for Image
Processing on Blackfin? Processors, Analog Devices, September, 2005.
[7] C. Lam, Using the ADSP-BF561 Blackfin? Processor as a TFT-LCD Controller,
Analog Devices, March, 2006.