透過您的圖書館登入
IP:18.191.170.239
  • 學位論文

以FPGA為基礎之即時車牌定位系統

FPGA-Based Real Time License Plate Localization System

指導教授 : 陳世旺
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


自動車牌辨識系統的研究發展也有數十年的歷史,國內、外皆有相當多的研究人員投入相關研究,也有相當多的研究成果技術產生,但是自動車牌辨識卻在人類的生活中無法普及化,其最大的原因是影像處理需要花費大量計算時間,所以導致車牌定位不夠快速和精確。若車牌辨識系統能夠具備高度的辨識率和可靠性,則可應用在無人監控的交通運輸系統上。本論文提出一個硬體架構之車牌定位系統,來加速其處理速度。 在車牌定位部分,主要是利用車牌上的色彩作為特徵,以偵測國內4種類型的車牌,主要是將非車牌顏色邊線去除,因為被保留下的車牌顏色邊線會較於集中,在利用形態學中的閉合、斷開、擴張,可將車牌邊線連成一塊,同時也將雜訊去除,再利用連通元件標籤法後,配合車牌長寬比和面積大小,可以偵測出車牌在影像中的位置。實驗結果中,利用FPGA來建構硬體,一張彩色影像的車牌定位的時間約6.542ms。

並列摘要


The study of license plate recognition (LPR) has been developed for over the decade of years. There were also plenty of contributions by this area of research. However, LPR is not the universal system in human daily life. The reason is that image processing costs large amount of computation by personal computer so that license plate does not preciously and slowly locate. If license plate location can be implemented by hardware, it will enhance the performance, speed, preciseness and reliability for the use of the system of intelligence transportation. The thesis proposed an approach of license plate location implemented by hardware in order to improve the speed of operation. In section of license plate location, using the color attribute of license plate is to be the feature for detecting four kinds of license plate in Taiwan, that is to eliminate the edge which is not belong to the color of license plate by color edge detection. Because the remained license- plate edges are almost closed to each other, the approach of morphology using closing, opening and dilation respectively can connect those edges to be a bigger region and remove the noise out simultaneously. After binarization, applying connect component labeling is to assign the unique number to each region. The size filter and aspect ratio can seize the location of license plate. In experimental results, the execution time of one color image is about 6.543ms by FPGA architecture, Color edge detection

參考文獻


[Yang 05] Feng Yang, Zheng Ma, “Vehicle License Plate Location Based on Histogramming and Mathematical Morphology”, Proceeding of the 4th IEEE Workshop on Automatic Identification Advanced Technologies, pp.89 – 94, 17-28 Oct. 2005.
[Bai 04] Bai Hongliang and Liu Changping,“A hybrid License Plate Extraction Method Based On Edge Statistics and Morphology”, Proceedings of the 17th International Conference on Pattern Recognition, Volume 2, PP.831-834, 23-26 Aug. 2004.
[Chang 04] 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, No.1, pp. 42-53, March 2004.
[Gon 02] Rafael C. Gonzalez, Richard E. Woods, “Digital Image Processing International Second Edition,” 2002.
[Urb 08] Erik R. Urbach, and Micheal H. F. Wilkinson, “Efficient 2D Grayscale Morphological transformations with arbitrary flat structuring elements,” IEEE Transactions on image processing, Vol. 17, No.1, January 2008.

延伸閱讀