臺灣博碩士論文加值系統

對於能改善行車安全的先進駕駛輔助系統，本論文提出基於光點偵測技術的夜間車輛辨識與追蹤系統，透過車燈的辨識追蹤車輛的位置。

具體而言，我們採用低曝光的方法來偵測來車的車頭燈，而在遠距離的前車車尾燈辨識部分，則是利用RGB影像空間中的紅色通道加上感興趣區間，之後再透過水平翻轉技術測量車燈的相似度，並將相似度高的配對為一組車燈對，以提高辨識的正確率。尤其，我們也考慮到其他車輛、路燈以及招牌等障礙物可能會擋住所考慮的車燈，於是運用卡爾曼濾波追蹤車燈以免遺漏，並使用匈牙利演算法找出前後幅影像中車燈的關聯性，以提高追蹤的正確率。根據模擬實驗的結果，我們所提出的方法不管是在高速公路或是一般道路，都有極高的辨識率，也能將市區許多光源的干擾排除。

In this thesis, for the Advanced Driver Assistance System (ADAS), which is able to improve driving safety, we propose a system for vehicle detection and tracking at nighttime with the light blob detection technique to identify and track a vehicle by recognizing its lamps.
Specifically, we detect the headlights of crossing vehicles in the way of low exposure setting. As for recognizing the taillights of distant lead vehicles, we first exploit the RGB color space with the red channel only as well as establishing the region of interest, and then measure the similarity between lamps by the horizontal flip technique to have each lamp image paired with the one with the highest similarity for improving the accuracy of detection. In particular, we consider that obstacles such as other vehicles, streetlamps, and shop signs may block vehicle lamps, and so on. Hence, we use the Kalman filter to track lamps for not missing any one, and find the correlation between the past and current frames with Hungarian algorithm to improve the accuracy of tracking. According to the results of simulation experiments, the method proposed by us not only can achieve a high recognition rate on both highways and ordinary roads, but also can eliminate much interference from lights in urban areas.

摘要 i
ABSTRACT ii
目錄 iii
圖目錄 iv
表目錄 vi
第一章 緒論 1
1.1 研究動機 1
1.2 論文架構 1
第二章 文獻探討 2
2.1 夜間車輛辨識 2
2.2 夜間車輛追蹤 4
第三章 研究方法 5
3.1 系統架構 5
3.2 夜間前方車輛與來車辨識 7
3.2.1 車燈辨識分類 7
3.2.2 影像二值化 9
3.2.3 感興趣區間（Region Of Interest） 12
3.2.4 積分影像 14
3.2.5 Star 偵測器（StarDetector） 16
3.2.6 車燈配對 17
3.3 夜間前方車輛與來車追蹤 21
3.3.1 卡爾曼濾波（Kalman filter） 21
3.3.2 匈牙利演算法（Hungarian algorithm） 24
第四章 實驗結果與討論 26
4.1 車輛辨識 28
4.2 演算法效能討論 34
第五章 結論與未來工作 36
參考文獻 37

[1]R. O’Malley, M. Glavin, E. Jones, “Vision-Based Detection and Tracking of Vehicles to The Rear with Perspective Correction in Low-Light Conditions,” IET Intell. Transp. Syst., Vol. 5, Iss. 1, pp. 1–10, 2011
[2]Sungmin Eum, Ho Gi Jung, “Enhancing Light Blob Detection for Intelligent Headlight Control Using Lane Detection,” IEEE Trans. Intell. Transp. Syst., Vol. 14, No. 2, pp. 1003–1011, June 2013.
[3]Naoya Kosaka, Gosuke Ohashi, “Vision-Based Nighttime Vehicle Detection Using CenSurE and SVM,” IEEE Trans. Intell. Transp. Syst., Vol. 16, No. 5, Oct 2015
[4]Qi Zou, Haibin Ling, Siwei Luo, Yaping Huang, Mei Tian, “Robust Nighttime Vehicle Detection by Tracking and Grouping Headlights,” IEEE Trans. Intell. Transp. Syst., Vol. 16, No. 5, Oct 2015
[5]Bima Sahbani, Widyawardana Adiprawita, “Kalman Filter and Iterative-Hungarian Algorithm Implementation for Low Complexity Point Tracking as Part of FastMultiple Object Tracking System,” IEEE Int. Conf. Syst. Eng. Tech., pp. 109-115, Oct 2016
[6]M. Agrawal, K. Konolige, and M. R. Blas, “CenSurE: Center surround extremas for real time feature detection and matching,” Lecture Notes Comput. Sci., vol. 5305, pp. 102–115, 2008.
[7]Z. Sun and G. Bebis, “Monocular precrash vehicle detection: Features and classifiers,” IEEE Trans. Image Process., vol. 15, no. 7, pp. 2019–2034, Jul. 2006.
[8]D. Alonso, L. Salgado, and M. Nieto, “Robust vehicle detection through multidimensional classification for on board video based systems,” in Proc. IEEE Int. Conf. Image Process., 2007, vol. 4, pp. 321–324.
[9]J. Arrospide and L. Salgado, “Log-Gabor filters for image-based vehicle verification,” IEEE Trans. Image Process., vol. 22, no. 6, pp. 2286–2295, Jun. 2013.
[10]M. Betke, E. Haritaoglu, and L. Davis, “Real-time multiple vehicle detection and tracking from a moving vehicle,” Mach. Vis. Appl., vol. 12, no. 2, pp. 69–83, 2000.
[11]S. Sivaraman and M. M. Trivedi, “General active learning framework for on-road vehicle recognition and tracking,” IEEE Trans. Intell. Transp. Syst., vol. 11, no. 2, pp. 267–276, Jun. 2010.
[12]Q. Yuan, A. Thangali, V. Ablavsky, and S. Sclaroff, “Learning a family of detectors via multiplicative kernels,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 33, no. 3, pp. 514–530, Mar. 2011.
[13]A. Lpez et al., “Temporal coherence analysis for intelligent headlight control,” in Proc. 2nd Workshop Planning, Perception Navigat. Intell. Veh., Nice, France, 2008, pp. 59–64.
[14]P. F. Alcantarilla et al., “Automatic light beam controller for drive assistance,” Mach. Vis. Appl., vol. 22, no. 5, pp. 819–835, Sep. 2011.
[15]P. F. Alcantarilla et al., “Night time vehicle detection for driving assistance light beam controller,” in Proc. IEEE Intell. Veh. Symp., 2008, pp. 291–296.
[16]Y. L. Chen, “Nighttime vehicle light detection on a moving vehicle using image segmentation and analysis techniques,” WSEAS Trans. Comput., vol. 8, no. 3, pp. 506–509, Mar. 2009.
[17]Y. L. Chen et al., “Vision-based driver nighttime assistance and surveillance system based on intelligent image sensing techniques and a heterogamous dual-core embedded system architecture,” Sensors, vol. 12, no. 3, pp. 2373–2399, 2012.
[18]J. Wang, X. Sun, and J. Guo, “Region tracking-based vehicle detection algorithm in nighttime traffic scenes,” Sensors, vol. 13, no. 12, pp. 16474–16493, 2013.
[19]H. Kim, S. Kuk, M. Kim, and H. Jung, “An effective method of head lamp and tail lamp recognition for night time vehicle detection,” in Proc. Int. Conf. Comput., Electr., Syst. Sci., Eng., Hong Kong, 2010, pp. 54–57.
[20]J. Rebut, B. Bradai, J. Moizard, and A. Charpentier, “Monocular vision based advanced lighting automation system for driving assistance,” in Proc. IEEE Int. Symp. Ind. Electron., 2009, pp. 311–316.
[21]R. O’Malley, E. Jones, and M. Glavin, “Rear-lamp vehicle detection and tracking in low-exposure color video for night conditions,” IEEE Trans. Intell. Transp. Syst., vol. 11, no. 2, pp. 453–462, Jun. 2010.