應用機器視覺技術於行人穿越道線與行人專用號誌辨識之研究

A Study of Pedestrian Crossing and Signal Recognition using Machine Vision

梁晉瑋(Jin-Wei Liang)

指導教授 ： 吳明川

國立臺北科技大學/機電學院/機電整合研究所/碩士(2010年)

https://doi.org/10.6841/NTUT.2010.00213

摘要

由於視障者在穿越道路時，常以車輛的引擎聲與盲人音響號誌做為判斷的依據，但是因為道路口常有喇叭和行人的聲音干擾，容易造成視障者的誤判，因此本研究利用機器視覺技術發展一套視障者輔助系統，進行行人穿越道線與行人專用號誌的辨識，提供道路的資訊來幫助視障者安全穿越道路。
本研究首先將RGB色彩空間轉為HSV，並且利用高斯混合模型偵測出行人專用號誌與行人穿越道線的色彩。再來利用區域標記偵測出可能位置，最後使用動態輪廓模型與相關係數法達成行人穿越道線偵測和行人專用號誌辨識的目的。經由實驗的結果可知，行人穿越道線在不同光線和視線與道路夾角 ﾟ的情況之下以及行人專用號誌在不同環境之下仍可辨識出來。

關鍵字

視障者 ； 行人穿越道線 ； 行人專用號誌

並列摘要

Because of visually handicapped pass through the road, they often use vehicles' sound and Audible pedestrian signal as the judgment, but because loudspeaker and pedestrian's sound’s disturbance, it’s easy to affect visually handicapped judgment, therefore this paper uses machine vision to develop a auxiliary system for the visually handicapped to recognize Pedestrian Crossing and Pedestrian Signal, providing road’s information to help visually handicapped pass through the road safely.
This paper first transfers RGB color model to HSV color model, and uses Gaussian mixture model to detect Pedestrian Crossing and Pedestrian Signal’s color, and then uses region mark to detect possible location, finally uses Snake model and Correlation Coefficient to recognize Pedestrian Crossing and Pedestrian Signal, by Experimental results we know the system can recognize when Pedestrian Crossing have ﾟ revolving and Pedestrian Signal in different environments.

並列關鍵字

Visually Handicapped ； Pedestrian Crossing ； Pedestrian Signal

參考文獻

[1] Bruno Andò, “Electronic Sensory Systems for the Visually Impaired,” IEEE Instrumentation & Measurement Magazine, 2003.

[3] A. D. L. Escalera, E. Moreno, M. A. Salichs, and J. M. Armingol, “Road Traffic Sign Detection and Classification,” IEEE Transactions on Industrial Electronics, vol. 44, 1997.

[5] 曹舜賢，視障者人行交通號誌辨識輔具之研發，碩士論文，國立台灣大學電機工程學系，台北，2007。

[6] Tadayoshi Shioyama, Haiyuan Wu, Naoki Nakamura and Suguru Kitawaki, “Measurement of the lenth of pedestrian crossings and detection of traffic lights form image data,” Measurement Science and Technology, 2002, pp1440–1447.

[7] Xiaoli Yang, Zhenpeng Zhao and Youn K Kim, “A Real Time Traffic Light Recognition System,” International Journal of Information Acquisition, 2008, pp149–161.