A tire identification system by using the image processing methods 林昭男、吳建達
E-mail: [email protected]
ABSTRACT
The goal of this research is to develop a skid-mark identification system that can automatically identify the skid-mark belongs of the vehicles at accident scene. That is, the system developed by using the image processing can segment tire-mark from the pictures and search the numbers and widths of the light and heavy striations on tire-mark for doing tire-tread matching, then find out these tire-marks are made by which cars at accident scene. This system uses some image-processing techniques such as binary, Sobel filter, thinning, Hough transform, rotation, horizontal axle projection etc. The operational procedure of the system has three steps to identify the tire-mark. The first step is pre-processing, Using Sobel filter, thinning and Hough transform to find out the oblique angle of the skid-mark. The second step is segment and taking the feature. Firstly, rotate that the skid-mark picture to the vertical direction.
Then, use the horizontal axle projection method to segment the skid-mark area from the picture. Final, combine Sobel filter and Hough transform to take the skid-mark’s feature that is amounts and widths of the light and heavy striations on skid-mark. The final step is using the skid-mark’s feature and the widths on tire-tread in data-base to do the template matching by Euclidean distance and find out each template’s error. If the error is smaller, the template is more similarity and the tire-tread is more possible to make the skid-mark.
Keywords : Words：Accident scene, tire-mark, image processing.
Table of Contents
封面內頁 簽名頁 授權書...iii 中文摘要...v 英文摘
要...vi 誌謝...vii 目錄...viii 圖目 錄...x 表目錄...xiii 符號說明...xiv 第 一章 序論 1.1 緣起...1 1.2 本文目標...1 1.3 文獻回
顧...2 1.4 系統之架構及進行步驟...5 1.5 論文架
構...7 第二章 相關理論探討 2.1 二值化...10 2.2 平滑處
理...10 2.3 膨脹處理...11 2.4 索貝濾波器...12 2.5 細線化...13 2.6 霍氏轉換...18 2.7 區域成長
法...20 2.8 旋轉...21 2.9 樣板匹配方法...22 2.10 投影法...23 第三章 前處理與胎痕切割 3.1 平滑處理...26 3.2 胎痕 梯度方向偵測...28 3.3 梯度大小偵測...31 3.4 細線
化...32 3.5 霍氏轉換...34 3.6 旋轉...44 3.7 二值化處理...45 3.8 水平投影法...47 第四章 胎痕特徵擷取與胎痕比對 4.1 胎痕特徵擷取...50 4.2 胎痕比對...56 第五章 結論 5.1 實驗結
果...57 5.2 結論...65 參考文獻...66 REFERENCES
[1] Y. W. Wang, “A Distance-base Matching Model for Classifying the Tire-marks at Accident Scene,” Journal of the Eastern Asia for Transportation studies, Vol. 5, pp. 2708-2721, 2003.
[2] S. Arnould, G. J. Awcock and R. Thomas, “Remote Bar-code localization using mathematical morphology,” Proceedings of IEE Seventh International Conference on Image processing and Its Applications, Vol. 2, pp. 642-646, 1999.
[3] A. K. Jain and Y. Chen, “Bar code localization using texture analysis,” Proceedings of the Second IEEE International Conference on Document Analysis and Recognition, pp. 41-44,1993.
[4] S. J. Liu, H. Y. Liao, Y. Chen, L. H Chen, H. R. Tyan and J. W. Sieh, “Camera-based bar code recognition system using neural net,”
proceedings of IEEE International Joint Conference on Neural Networks, Vol. 2, pp. 1301-1305, 1993.
[5] R. Muniz, L. Junco and A. Otero, “A robust software barcode reader using the Hough transform,” Proceedings of IEEE International Conference of Information Intelligence and Systems, pp. 313-319, 1999.
[6] 劉松益，“以梯度運算子為基礎的條碼定位技術”，大同大學 資訊工程研究所碩士論文，2000。
[7] R. C. Gonzalez and R. E. Woods, “Digital Image Processing,” Addison-Wesley Publishing Company, 1992.
[8] C. L. Su, “Face recognition by feature orientation and feature geometry matching,” Ph.D. dissertation, The University of Southwestern Louisiana, 1995.
[9] 王瑩瑋、吳建達、林昭男，“影像處理技術於事故現場胎痕 鑑定上之運用”，道路交通安全與執行法研討會論文集，2003。
[10] 吳成柯、程湘君、戴善榮、雲立實，“數位影像處理”，儒林 圖書有限公司，1996。
[11] 林宸生，“數位信號影像與語音處理”，全華科技圖書股份有 限公司，1998.
