近幾年來，層出不窮的車輛失竊案件，已為民眾所擔憂之問題。針對車輛失竊問題，警方雖有其處理方法，但效果與執行效率均不佳。為了改善警方之偵辦效率，本研究利用電腦視覺技術，提出自動贓車偵測系統，其結合車牌辨識軟體、攝影機、GPS接收器，以及網路設備，其利用攝影機擷取車牌影像，再透過車牌辨識軟體，自動辨識其中之車牌號碼，改善警方目前以人為輸入車牌號碼的方式，減少操作的過失，並結合現有網路系統，提供即時更新的能力。本研究所提出之車牌辨識系統可分為車牌定位系統與文字辨識系統。在車牌定位系統方面，本研究利用total variation decomposition，改善影像之光影變化問題，並利用color edge detection與morphology operation，在影像內尋找可能為車牌之區塊(candidate)，並根據aspect ratio，去除不可能為車牌之區塊。在文字辨識系統方面，其將車牌定位之結果，進行文字分割處理，並透過skewness correction，改善文字區塊的變形現象。在文字辨識時，先根據文字區塊之拓樸性質，對其進行初步分類，減少文字比對的次數，並利Chamfer distance，計算文字區塊與標準字元之相似程度，並以相似成度較高者，作為辨識結果。由實驗結果知，即使在光影變化的環境，本研究之定位系統仍可定位車牌之位置，而針對歪斜之車牌，透過skewness correction之處理，可提升文字辨識系統的正確性。

In this thesis, an automatic stolen vehicle detection system is proposed. This system is composed of a PC-based host computer, a color CCD camera, a GPS device, and a communication unit. A license plate recognition (LPR) program and a database of stolen license numbers have been installed in the computer. The CCD camera captures images as the input data to the LPR program. Once an input license number is recognized as belonging to a stolen vehicle through comparing with those pre-stored in the database, the license number, GPS information, and image of the vehicle are immediately transmitted to a control center through the communication unit.
The license plate recognition program plays an important role in the proposed system, which consists of two modules: the license plate location and the license number identification modules. In the license plate location module, the total variation method is first applied to the input image for normalizing its illumination. Thereafter, a color edge detector looks for in the resultant images particular edges that may be related to license plates. Morphology operations are then applied to the edge image, which highlight the image areas that may contain license plates. Only the areas whose aspect ratios agree with that of license plates are preserved. In the license number identification module, for each license plate candidate, character segmentation is performed, followed by a skewness correction process, which normalizes the shapes of segmented characters. In the final step, characters are recognized to accomplish the license number identification.
A series of outdoor experiments have been conducted to demonstrate the feasibility of the proposed system.

[Bas00] R. Basri and D. Jacobs, “Lambertian Reflectance and Linear Subspaces,” Technical Report 2000-172R, NEC Research Inst., 2000.

[Bel96] P.N. Belhumeur and D.J. Kriegman, “What Is the Set of Images of an Object under All Possible Lighting Conditions?” IEEE Int’l Conf. on Computer Vision and Pattern Recognition, Jun. 1996.

[Bro05] A. Broumandnia and M. Fathy, “Application Of Pattern Recognition For Farsi License Plate Recognition,” Int’l Journal of Graphics, Vis. Image Process, Vol. 2, pp. 25-31, Jan. 2005.

[Cha04] T. F. Chan and S. Esedoglu, “Aspects Of Total Variation Regularized L1 Function Approximation,” CAM Report 04-07, Univ. of California, Los Angeles, Feb. 2004.

[Che06] T. Chen, X. S. Zhou, D. Comaniciu, and T. S. Huang, “Total Variation Models for Variable Lighting Face Recognition,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 28, No. 9, Sep. 2006.

[Dua04] T. D. Duan, D. A. Duc and T. L. H. Du, “Combining Hough Transform and Contour Algorithm for Detecting Vehicles' License-Plate,” Proc. of Int’l Symp. on Intelligent Multimedia, Vedio and Speech Processing, pp. 747-750, Oct. 2004.

[Fou03] C. Fouard, “Automatic calculation of chamfer mask coefficients for large masks and anisotropic images,” INRIA, pp. 7 -10, 2003.

[Gao07] Q. Gao, X. Wang and G. Xie, “License Plate Recognition Based On Prior Knowledge,” IEEE Int’l Conf. on Automation and Logistics, pp. 2964-2968, Aug. 2007.

[Guo08] J. M. Guo, and Y. F Liu “License Plate Localization and Character Segmentation With Feedback Self-Learning and Hybrid Binarization Techniques,” IEEE Trans. on Vehicular Technology, Vol. 57, No. 3, pp. 1417-1424, May 2008.

[Jun05] R. Juntanasub and N. Sureerattanan, “Car License Plate Recognition through Hausdorff Distance Technique,” IEEE 17th Int’l Conf. on Tools with Artificial Intelligence, Nov. 2005.

[Kim00] K. K. Kim, K. I. Kim, J. B. Kim, and H. J. Kim, “Learning-based approach, for license plate recognition,” Proc. of IEEE on Signal Processing Society Workshop , Vol. 2, pp. 614–623, Dec. 2000.

[Lu08] Shuxian Lu, Zhengxi Liu, Yantao Chen, Lin Liu “AWHD for License Plate Character,” IEEE Int’l Conf. on Embedded Software and Systems Symposia, pp. 60-62, Jul. 2008.

[Ots79] N. Otsu, “A Tlreshold Selection Method from Gray-Level Histograms” IEEE Trans. on Systems, Man, and Cybernetics, Vol. 9, No. 1, Jan. 1979.

[Per90] P. Perona and J. Malik, “Scale-Space and Edge Detection Using Anisotropic Diffusion,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 12, no. 7, pp. 629-639, Jul. 1990.

[QIN 06] Z. Qin, S. Shi, J. Xu, and H. Fu, “Method of License Plate Location Based on Corner Feature,” IEEE Congress on World Congress on Intelligent Control and Automation, pp. 8645-8649, June 2006.

[Rud92] L. Rudin, S. Osher, and E. Fatemi, “Nonlinear Total Variation Based Noise Removal Algorithms,” Physica D, vol. 60, pp. 259-268, 1992.

[Sar07] M. Sarfraz, S.A. Mahmoud and Z. Rasheed, “On Skew Estimation and Correction of Text,” Proc. of IEEE on Computer Graphics, Imaging and Visualisation, pp. 308-313, Aug. 2007.

[Sha01] A. Shashua and T. Riklin-Raviv, “The Quotient Image: Class-Based Re-Rendering and Recognition with Varying Illuminations,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 23, no. 2, pp. 129-139, Feb. 2001.

[Son99] M. Sonka, V. Hlavac, and R. Boyle, Image Processing, Analysis, and Machine Vision, PWS, 1999.

[Sto72] T.G. Stockham, “Image Processing in the Context of a Visual Model,” Proc. of IEEE, vol. 60, pp. 828-842, Jul. 1972.

[Tan05] S. T. Tang and W. J. Li, “Number and Letter Character Recognition of Vehicle License Plate Based on Edge Hausdorff Distance,” IEEE 6th Int’l Conf. on Parallel and Distributed Computing, pp. 850-852, Dec. 2005

[Vio04] P. Viola and M. J. Jones, “Robust Real-Time Face Detection,” Int’l Journal of Computer Vision, Vol. 57, No. 2, pp. 137–154, May 2004.

[Wen01] C. Y. Wen, C. C. Yu and Z. D. Hun, “Normalization of a Skew Distoration Vehicle License Plate image Based on Three-Dimensional Transformation Techniques,” IPPR Conf. on Computer Vision, Graphics, and Image Processing, Aug. 2001.

[Yan05] F. Yang, Z. Ma, “Vehicle License Plate Location Based On Histogramming And Mathematical Morphology,” IEEE 4th workshop on Automatic Identification Advanced Technologies, pp 89-94, Oct. 2005.

[Yan07] F. Yang, Z. Ma and M. Xie, “A robust character segmentation approach for license plate,” IEEE Int’l Conf. on Communications, Circuits and Systems, pp. 679-683, Jul. 2007.

[Zha06] H. Zhang and W. Jia, “A Fast Algorithm for License Plate Detection in Various Conditions,” IEEE Int’l Conf. on Systems, Man, and Cybernetics, pp. 2420-2425, Oct. 2006.

[Zha84] T. Y. Zhang and C. Y. Suen, “A Fast Parallel Algorithm for Thinning Digital Patterns,” Communications of the ACM, Vol. 27, pp. 236-239, Mar. 1984.