摘要
近年來隨著數位影像處理的車牌定位與車速偵測發展蓬勃, 應用層面也從一 般道路延伸至收費站、停車場等場所之車輛控管。 本研究提出了使用定位後之車 牌來偵測車速, 以避免偵測車輛超速後還必須額外進行車牌定位, 其速度與效能 比傳統偵測方式更快速。 無論車輛超速與否, 其車牌皆會切割並儲存於資料庫, 可更有效實現車輛監控。 本研究提出之偵測車速方式可減少攝影機所需要校正之 參數。 一般的測速方式主要在拍攝的影像裡畫上兩條校準線, 或基於影像之透 視效果所以必須使用投影在地面的車輛資訊來估算車速。 本研究的距離估算目標 物不需要緊貼地面, 並將車體顏色區分為淺色與深色兩大類別, 進行自動全時 段的車輛進入偵測、車牌偵測、車牌追蹤與車速計算。 實驗結果之整體成功率達 95%, 且車速結果與 GPS 量測車速之誤差小於 2km/h。
並列摘要
Recently, as the development of digital image processing of plate-locating and speed-detecting, the application of plate-locating and speed-detecting for vehicle con- trol expands from streets to toll stations and parking lots. Our research proposes detecting the speed of vehicles by located plates which is faster than traditional detecting ways. In this way, there is no need to locate the plates for the speeding vehicles. The located plates will be saved in database regardless of the speeds of the vehicles which could supervise vehicles more effectively. The speed calculation method we propose reduces the camera parameters which need to be calibrated. There are mainly two speed-detecting methods. One way is to use two calibration lines within the images. Considering of the perspective of the images, the other is to use the information of the vehicle which is projected on the ground. However, in our research, the objects in the images which are going to be esti- mated the distance to the camera don’t have to be projected on the ground. In addition, we distinguish colors of vehicles into light color and dark color to process vehicle detecting, plate detecting, plate tracking and speed estimation in all day automatically. The correc- tion rate of overall system is 95% and the estimated error of the vehicle speed is under 2 km/h.
參考文獻
[24] 李濠欣, “即時相機校正之前車偵測,” 國立台灣大學工程科學及海洋工程學系 碩士論文, 2009.
[5] X. C. He and N. Yung, “A novel algorithm for estimating vehicle speed from two consecutive images,” in IEEE Workshop on Applications of Computer Vision, feb. 2007, pp. 12 –12.
[6] F.Yamazaki,W.Liu,andT.Vu,“Vehicleextractionandspeeddetectionfromdigital aerial images,” in IEEE International Geoscience and Remote Sensing Symposium, vol. 3, 7-11 2008, pp. 1334 –1337.
[7] M. Litzenberger, A. Belbachir, N. Donath, G. Gritsch, H. Garn, B. Kohn, C. Posch, and S. Schraml, “Estimation of vehicle speed based on asynchronous data from a silicon retina optical sensor,” in IEEE Conference on Intelligent Transportation Systems, 17-20 2006, pp. 653 –658.
[8] V. Madasu and M. Hanmandlu, “Estimation of vehicle speed by motion tracking on image sequences,” in IEEE Intelligent Vehicles Symposium, june 2010, pp. 185 –190.