臺灣博碩士論文加值系統

臺灣目前是全世界經濟高度發展的國家之一，國內的車流量更是一年比一年還要多，對於車流量監控系統的需求也就日益增加，而夜間車流監控系統也是近年來車流量監控系統上一個非常重要的主題，在影像處理學術界上也是一個新興的領域；國道大量使用智慧型運輸系統(intelligent transportation system, ITS)來監控交通量，而國道的監視器因為要做長時間監控，且要對夜間環境進行監控，所以都配備紅外線功能，極為昂貴。是以本研究希望能透過智慧型運輸系統結合一般監視器材，以夜間環境與國道為主，來進行夜間車流分析。本研究以RGB值捕捉燈源並判定該燈源是否有移動特性，來判斷該光源是否為車燈，每個車輛至少有兩個車頭燈，故必須將捕捉到之燈源做分群，來判斷被分群之燈源為一台車，以此計數。本研究以國道為主拍攝了15段道路監控影片，在本研究的結果顯示，監視器材上成功降低將近75%的成本，在車流量偵測率上也有91%的正確率。

Taiwan is currently the highly developed countries in the world, the traffic flow is even more than last year, the demand of traffic monitoring system is also increasing, in recent years, nighttime traffic monitoring system is a very important issue in traffic monitoring system is a very important theme, and it’s also an emerging academic field in the image processing; freeway monitoring use the intelligent transportation systems (intelligent transportation system, ITS) to monitor traffic flow, and monitor freeway require a long time monitoring, and monitoring in the night, the equipped with infrared capabilities, extremely expensive. This study is based on through intelligent transportation systems combining general surveillance equipment, and focus on monitoring traffic flow on freeway in nighttime for traffic analysis. In this study, we use the RGB values to capture the light source in the image and detecting the headlights of a vehicle depend it’s moving or not. in order to count, each vehicle has at least two headlights, it must be grouping two headlights then can detecting the vehicle. In this study, we shot 15 films on freeway, in this study showed that surveillance equipment to successfully reduce nearly 75% of the cost, and the vehicle detection has 91% accuracy.

中文摘要 i
英文摘要 ii
致 謝 辭 iii
目錄 iv
圖目錄 vi
表目錄 vii
第一章　緒論 1
第一節　研究背景 1
第二節　研究動機 3
第三節　研究方法 4
第四節　研究流程 5
第五節　論文架構 6
第二章　文獻探討 7
第一節　智慧型運輸系統 7
第二節　車燈辨識 8
第三節　車燈分群 9
第四節　車流分析 10
第三章　車燈捕捉與配對 11
第一節　車燈的捕捉與追蹤 11
第二節　車燈配對 16
第四章　車流分析 19
第五章　實驗結果 21
第一節　結果分析 21
第六章　結論 24
參考文獻 26

一、中文部分

[1]交通部(2013)，臺灣地區機動車輛登記用數[線上資料]，來 源:http://www.dgbas.gov.tw/mp.asp?mp=1[2012, dec 31]

[2]交通部台灣區國道高速公路局(2007)，智慧型運輸系統ITS簡介[線上資料]，來源：http://www.freeway.gov.tw/Publish.aspx?cnid=1556 [2011, feb 8]

[3]陳華總，蔡立武，古蕙媜，李素瑛，林寶樹(2012)，利用車燈偵測分析夜間交通壅塞情況，前瞻科技與管理期刊，2(1)，67-79。

二、英文部分

[4]Congcong Li et al.,2011, “Joint Optimization of Background Subtraction and Object Detection for Night Surveillance”, IEEE International Conference on Image Processing.

[5]Chen YL et al., 2009, “Real-Time Vision-Based Multiple Vehicle Detection and Tracking for Nighttime Traffic Surveillance,” in Proceedings of IEEE International Conference on Systems, Man, and Cybernetics, San Antonio, TX, US: Institute of Electrical and Electronics Engineers, 3352-3358.

[6]Hsieh JW et al.,2006, “Automatic Traffic Surveillance System for Vehicle Tracking and Classification”, IEEE Transaction on Intelligent Transportation Systems, 7, 175 187.

[7]Huang K et al., 2008, “A Real-Time Object Detecting and Tracking System for Outdoor Night Surveillance,” Pattern Recognition, 41, 432-444.

[8]Lee W and Ran B,2006, “Bidirectional Road way Detection for Traffic Surveillance Using Online CCTV Videos,” in Proceedings of IEEE International Conference on Intelligent Transportation Systems, Toronto, Canada: Institute of Electrical and Electronics Engineers, 1556-1561.

[9]Li B and Chen QM, 2009, “Framework for Freeway Auto-Surveillance from Traffic Video,” in Proceedings of WRI World Congress on Computer Science and Information Engineering, Los Angeles, CA, US: Institute of Electrical and Electronics Engineers, 360-365.

[10]Rao B et al., 1993, “A Fully Decentralized Multi-Sensor System for Tracking and Surveillance,” The International Journal of Robotics Research, 12, 20-44.

[11]Robe r t K, 2009, “Night -Time Traffic Surveillance: A Robust Framework for Multi-Vehicle Detection, Classification and Tracking,” in Proceedings of IEEE International Conference on Advanced Video and Signal Based Surveillance, Genova, Italy: Institute of Electrical and Electronics Engineers, 1-6.

[12]Sayed MS and Delva J, 2010, “Low Complexity Contrast Enhancement Algorithm for Nighttime Visual Surveillance,” in Proceedings of the 10th International Conference on Intelligent Systems Design and Applications, Cairo, Egypt: Institute of Electrical and Electronics Engineers, 835-838.

[13]Tsai LW et al., 2011, “Multi-Lane Detection and Road Traffic Congestion Classification for Intelligent Transportation System,” Energy Procedia, 13, 3174-3182.

[14]Tseng BL et al., 2002, “Real-Time Video Surveillance for Traffic Monitoring Using Virtual Line Analysis,” in Proceedings of the IEEE International Conference on Multimedia and Expo, Lausanne, Switzerland: Institute of Electrical and Electronics Engineers, 541-544.

[15]Wei Zhang et al., 2012, “Tracking and Pairing Vehicle Headlight in Night Scenes, ” IEEE Transactions on Intelligent Transportation Systems, 13(1), 140-153.

[16]Yen-Lin Chen et al.,2011, “ A Real-Time Vision System for Nighttime Vehicle Detection and Traffic Surveillance, ” IEEE Transactions on Industrial Electronics, 58(5), 2030-2044.

[17]Wu J et al., 2007, “Virtual Line Group Based Video Vehicle Detection Algorithm Utilizing Both Luminance and Chrominance,” in Proceedings of 2nd IEEE Conference Industrial Electronics and Applications, Harbin, China: Institute of Electrical and Electronics Engineers, 2854-2858.