在本論文中，主要的目的是建置一個分散式平行車牌辨識系統於一群擁有多核心處理器的異質性計算節點上。在系統架構設計方面，主要分成四個部份，第一部份為「影像輸入模組」，第二部份為「負載分配模組」，第三部份為「車牌辨識模組」，第四部份為「車牌資訊管理模組」。
在程式架構方面，異質性計算節點本身，主要是使用多執行緒的概念，配合OpenMP與PThread函式庫來實作平行處理的功能，達成最大化系統資源使用的目標。在異質性計算節點通訊方面，主要是藉由訊息傳遞介面標準(MPI)函式庫來實作資料傳輸的功能，達成資料交換目的。在資料儲存與管理方面，主要是使用MySQL資料庫來儲存辨識的結果，並且搭配動態網頁技術(PHP)來建置圖形化介面，方便使用者瀏覽與查詢辨識的結果。
最後在實驗結果部份，分別對循序處理程式與平行處理程式進行效能的分析與比較，來證明本論文所提出的平行處理架構，在整體處理效能方面，比循序處理程式有更好的表現。

The main purpose of this thesis is to build a distributed and parallel license plate recognition system on multi-core and heterogeneous computing environments. The architecture of this system can be divided into four parts. The first part is "Video Input Module", the second part is "Work Load Distribution Module ", the third part is "License Plate Recognition Module", and the last part is "License Plate Information Management Module".
For maximizing the use of total system resources, it uses two libraries, OpenMP and PThread, to implement parallel processing capabilities. And the communications between computers is achieved by the standard message passing interface (MPI) library. For the convenience of users browsing and querying recognition results, two methods are used for storing and managing data. First, it uses MySQL database to store the recognition results, and then uses dynamic web technologies (PHP) to build a graphical interface.
In order to analyze the execution performance between series program and parallel program, we experiment different configurations of input video streams to our system. The result shows that the parallel processing architecture system in this thesis has better performance than serial program. It also identifies how to decompose application to gain the speed up.
Keyword: Parallel Computing, MPI, OpenMP, PThread, Load Balancing,
License Plate Recognition System

[1] Gabriele Jost* and Haoqiang Jin, “Comparing the OpenMP, MPI, and Hybrid Programming Paradigms on an SMP Cluster1”, NAS Technical Report NAS-03-019, NASA Ames Research Center, Moffett Field, CA , (2003)
[2] J. Corbal´an, A. Duran, J. Labarta “Dynamic Load Balancing of MPI+OpenMP applications”, ICPP04, Montreal, Quebec. Canada. (2004)
[3] K. Koyama, K. Shimizu, H. Ashihara, Y. Zhang, and H. Kameda, “Performance Evaluation of Adaptive Load Balancing Policies in Distributed Systems”, Proceedings of the Singapore International Conference on Networks/International Conference on Information Engineering '93, pp. 606-611 (1993)
[4] Blaise Barney, Introduction to Parallel Computing, Lawrence Livermore National Laboratory. [Online]. Available:
https://computing.llnl.gov/tutorials/parallel_comp/.
[5] Blaise Barney, Message Passing Interface, Lawrence Livermore National Laboratory. [Online]. Available: http://www.llnl.gov/computing/tutorials/mpi/.
[6] Blaise Barney, OpenMP, Lawrence Livermore National Laboratory. [Online]. Available: https://computing.llnl.gov/tutorials/openMP/.
[7] 王鵬，雲端運算的關鍵技術與應用實例，佳魁資訊股份有限公司，台北，3-30頁(2010)。
[8] S.P. Dandamudi, “Sensitivity Evaluation of Dynamic Load Sharing in Distributed Systems”, IEEE Concurrency, Vol. 6, No. 3, pp. 62-72 (1998)
[9] N. Fujimoto, K. Hagihara, “A Comparison among Grid Scheduling Algorithms for Independent Coarse-Grained Tasks”, Proceedings of SAINT 2004 Workshop on High Performance Grid Computing and Networking, pp. 674--680, Tokyo, Japan, January (2004)
[10] 謝志松、陳重臣，「系統模擬應用於異質分散式系統之動態平衡探討」，碩士論文，國立雲林科技大學，雲林(2003)。
[11] A. S. Banawan, N. M. Zeidat, “A Comparative Study of Load Sharing in Heterogeneous Multicomputer Systems”, Proceedings of the 25th annual symposium, pp. 22-31, April (1992)
[12] Shivaratri, N.G., Krugeger, P., and Singhal, M., “Load Distributing for Locally Distributed Systems”, Computer, Dec., Vol. 25,No. 12,pp. 33-44 (1992)
[13] W. Cai, B.S. Lee, A. Heng and L. Zhu, “A Simulation Study of Dynamic Load Balancing for Network-based Parallel Processing”, Proceedings of Third International Symposium on Parallel Architectures, Algorithms and Networks, pp.383-289, (1997).
[14] R. Elmasri and S. B. Navathe, Fundamentals of Database Systems, 4th ed., Addison Wesley,（2006）
[15] H.J. Siegel, H.G. Dietz, and J.K. Antonio, “Software Support for Heterogeneous Computing“, The Computer Science and Engineering Handbook, A.B. Tucker Jr., ed., pp. 1,886-1,909.Boca Raton, Fla.: CRC Press, (1997)
[16] 鄭守成，C語言MPI平行計算程式設計，國家高速網路與計算中心，台北(2002)。
[17] 李志文，「嵌入式即時多標的汽機車牌照辨識系統」，碩士論文，國立台灣科技大學，台北（2007）。
[18] 林明浩，「矩陣處理平行程式產生器之設計與實現」，碩士論文，國立台灣科技大學，台北(2010)。
[19] 徐志融，「基於手持行動裝置實現車牌資訊管理與辨識系統」，碩士論文，國立台灣科技大學，台北(2010)。