在現今的社會中，汽車在日常生活中已經成為我們不可或缺的一部分，雖然在市面上已存在許多汽車防盜設備，如方向盤鎖、指紋辨識系統等，但這些防盜設備都各有他們的缺點，導致在社會上還有許多汽車因而失竊，因此本論文針對汽車安全方面來進行研究。
本論文使用的實驗資料為12位駕駛者，透過駕駛一條城鎮道路而收集到的駕駛行為資訊，最後得到的資料為每位駕駛者各25筆資料，本論文使用生物特徵識別技術的駕駛行為來做識別，我們認為在正常的情形下，每個駕駛在開車時的身體行為會有一種固定的模式，因此本實驗透過兩架Kinect攝影機錄製使用者的駕駛行為，並透過IPi Motion Capture System商業軟體將使用者的身體骨架節點轉為行為資訊，並搭配分類器進而判別是否為合法使用者。
本論文提出一種新的防止汽車失竊的非侵入式驗證機制，透過直線路段及轉彎路段的實驗資料，每隔一段時間進行一次驗證，並在比較所有實驗結果後推薦最佳組合，而此最佳組合的EER效果可以達到9.5%。
本論文的EER效果可以達到9.5%，表示此非侵入式驗證機制可以解決現今人們因為覺得不方便而不使用的防盜系統問題，而且在未來此非侵入式驗證機制可以與其他驗證機制合併來更加降低失竊率，以增加汽車的安全性。
關鍵字:汽車安全、非侵入式使用者識別

Nowadays, car has become an indispensable part in our daily life. However, the increasing car-stolen cases bring an unignorable issue to the society. To suppress those criminal cases, lots of invented anti-theft devices such as wheel locks, alarm system and fingerprint identification systems are widely used in the world. Since some weakness still exist in these devices, the number of criminal cases are kept in a high level. Therefore, our study aims on the improvement of the vehicle security.
The dataset in our experiment is collected from twelve participants including eleven males and one female, and each has 25 samples of behavior information recorded when they drive on the urban road. The biometric identification technology is applied to our new authentication mechanism which assumes every person has his/her own fixed patterns when they are driving. Also, the Kinect camera is used to film the behavior of participants. The transformed information is then processed for the driver identification by using KNN and linear SVM, which then can automatically recognize the genuine user or the imposter.
More, a new authentication mechanism is proposed to raise the identification efficiency by additional verification from straight and curve road driving data. After tests of various combinations, we present a best combination which can reach 9.5% of EER. The result shows our new mechanism is promising in reinforcement for the anti-theft mechanism. In the future, the possible combination with other biometric mechanisms can be expected to reduce theft rate and also upgrade the vehicle protection.

[1] H. Song, S. Zhu, and G. Cao, “SVATS: A sensor-network-based VehicleAnti-Theft System,” in Proc. IEEE INFOCOM’08, Phoenix, AZ, USA, April 14-18, 2008
[2] Gamboa, H., and Fred, A. A behavioral biometric system based on human-computer interaction. In SPIE 5404 - Biometric Technology for Human Identification A. K. Jain and N. K. Ratha, Eds., 381–392
[3] Pusara and C. E. Brodley, "User re-authentication via mouse movements," in Proceedings of the 2004 ACM workshop on Visualization and data mining for computer security, 2004, pp. 1-8.
[4] L. O’Gorman,“Comparing passwords, tokens, and biometrics for user authentication,”Proc. IEEE, vol. 91, no. 12, pp. 2019–2040, Dec. 2003.
[5] D. Maio1, D.M., R. Cappelli1, J. L. Wayman2, A. K. Jain3, FVC2000: Fingerprint Verification Competition D. Maio, D. Maltoni, R. Cappelli, J.L. Wayman, and A.K. Jain,™FVC2000: Fingerprint Verification Competition,∫ Biolab internal report, Univ. of Bologna, Italy, Sept. 2000, available from http://bias.csr.unibo.it/fvc2000/..
[6] Huihuan Qian et al., "SupportVectorMachine for Behavior-Based Driver Identification System.", Journal of Robotics, Volume 2010 (2010), Article ID 397865, 11 pages.
[7] Ronghua Chen et al, "Driver Recognition Based on Dynamic Handgrip Pattern on Steeling Wheel.",Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing(SNPD), p107-112,Sydney, NSW, 6-8 July 2011.
[8] Yang, C.-H., A New Non-Intrusive Authentication Method based on Dynamics of Driver’s Upper Body Joint Angles. (to appear:The 12th annual IEEE consumer communications & networking conference, 2015)
[9] A. A. E. Ahmed and I. Traore, "A new biometric technology based on mouse dynamics," Dependable and Secure Computing, IEEE Transactions on, vol. 4, pp. 165-179, 2007.
[10] M. Pusara and C. E. Brodley, "User re-authentication via mouse movements," in Proceedings of the 2004 ACM workshop on Visualization and data mining for computer security, 2004, pp. 1-8.
[11] s.Cheng and M.Trivedi, "Turn-intent analysis using body pose for Iintelligent driver assistance,"IEEE Pervasive Computing, vol.5, no.4 pp. 28-37,2006.
[12] C. Tran and M. M. Trivedi, “Towards a vision-based system exploring 3-D driver posture dynamics for driver assistance: Issues and possibili-ties,” in Proc. IEEE Intell. Veh. Symp., 2010, pp. 179–184.
[13] Microsoft Inc. Microsoft kinect for windows SDK. Available from: http://en.wikipedia.org/wiki/Kinect
[14] City Car Driving. Car Simulator. Available from: http://citycardriving.com.
[15] Logitech Inc. G27 Racing Wheel. Available from: http://en.wikipedia.org/wiki/Logitech_G27 .
[16] iPi Soft. iPi Motion Capture™ Software. Available from: http://ipisoft.com.
[17] C. Cortes and V. Vapnik, "Support-vector networks," Machine learning, vol. 20, pp. 273-297, 1995.
[18] F.Lin, D.Liang, and E.Chen,“Financial ratio selection for businesscrisis prediction,”Expert Systems with Applications,vol.38, pp.15094-15102, 2011.
[19] Lin., C.-C.C.a.C.-J. A Library for Support Vector Machines.; Available from: http://www.csie.ntu.edu.tw/~cjlin/libsvm/.
[20] T. Fawcett, “An Introduction to ROC Analysis,”Pattern Recogni-tion Letters,vol. 27, no. 8, pp. 861-874, 2006.