在各品牌汽車系統中，近年來為發展智慧車輛以及無人自駕車，先進駕駛輔助系統(Advanced Driver Assistance Systems；ADAS)乃達到真正無人自動駕駛之過程中所需要的技術之一。
本論文利用個人電腦為上位控制器，試圖透過已訓練好之深度學習模型，透過影像偵測技術，找到紅綠燈及車輛在影像中的位置資訊，以完成無人駕駛時之自動辨識與偵測。依據前車距離遠近做為控制目標，以串列傳輸的方式利用PD控制器控制改裝電動車的油門、方向與剎車，而達到簡易自適應巡航控制。論文中以電腦視覺與影像處理開源函式庫(Open Source Computer Vision Library；OpenCV)進行影像處理與分析，完成紅綠燈之號誌辨識。
結果顯示以單一鏡頭完成辨識道路紅綠燈號誌及自適應巡航控制的可行性，以及在未來開發應用上的可拓展性。

In various brands of automotive systems, in recent years, in order to develop smart vehicles and self-driving cars, Advanced Driver Assistance Systems (ADAS) is one of the technologies required in the process of truly Self-Driving .
This thesis uses a personal computer as a host controller to try to find the location information of traffic lights and vehicles in the image through the well-trained deep learning model to complete the automatic identification and detection of Self-Driving.According to the distance of the front vehicle as the control target, the PD controller is used to control the throttle, direction and brake of the modified electric vehicle in a serial transmission mode to achieve simple adaptive cruise control.In this paper, the Open Source Computer Vision Library (OpenCV) is used for image processing and analysis to complete the identification of traffic lights.
The results show the feasibility of identifying road traffic lights and adaptive cruise control with a single lens, and the scalability of future development and application.

[1] 岩子庭，2016，以影像為基礎之日間前車偵測與車距估算，元智大學電機工程學系碩士論文。
[2] 莫又寧，2017，以混合式特徵偵測前方車尾並使用卷積神經網路辨識車型，元智大學電機工程學系碩士論文。
[3] 林奇叡，2008，先進安全車輛的前方與盲點視覺偵測，中央大學資訊工程學系碩士論文。
[4] 鄭弼倫，2011，高速公路前車陰影結合車輪之偵測，中正大學電機工程學系學位論文。
[5] Zehang Sun ,G. Bebis ,R. Miller “Monocular precrash vehicle detection: features and classifiers.” IEEE Transactions on Image Processing, Vol.15, Issue: 7, July 2006 2019-2034
[6] 葉冠甫，2012，基於單一影像之車輛偵測系統軟硬體設計，國立中山大學資訊工程學系研究所碩士論文。
[7] 楊道偉，2018，基於卷積神經網路之市區道路場景自適應車輛偵測機制，國立中央大學資訊工程學系碩士論文。
[8] 王鈞奕，2018，利用卷積神經網路對空拍影像中車輛進行偵測與分類，中原大學資訊工程研究所碩士論文。
[9] 劉朋宣，2017，基於深度學習之空拍影像車輛偵測技術，國立臺灣海洋大學電機工程學系碩士論文。
[10] 張豐安，2016，應用於先進駕駛輔助系統之多重物件偵測深度學習技術與其嵌入式平台實現，國立交通大學電資工程學系碩士論文。
[11] 曾慶鎧，2018，基於深度學習多重物件偵測與行為辨識之後方超車警示系統，國立交通大學電子研究所碩士論文。
[12] 戴瑋劭，2013，基於行車影像之危險駕駛偵測系統，國立交通大學資訊科學與工程研究所碩士論文。
[13] 陳政德，2016，基於單攝影機影像處理之前車距離估測，國立中正大學電機工程研究所碩士論文。
[14] Chao-Ho Chen, Tsong-Yi Chen, Deng-Yuan Hiang, and Kai-Wei Feng, "Front Vehicle Detection and Distance Estimation Using Single-Lens Video Camera," Proceedings of IEEEInternational Conference onRobotVision and Signal Processing(RVSP), Qingdao, China, Dec. 2012.

[15] Giseok Kim, Jae-Soo Cho, “Vision-based Vehicle Detection and Inter-Vehicle Distance Estimation”, Proceedings of IEEE International Conference on Control Automation and Systems (ICCAS), Jeju Island Korea, Oct. 2012.
[16] 趙子頎，淺談適應性巡航控制系統，車安通訊季刊，2018，第107-3期
[17] 陳建次，2006，新型全速度域模糊適應巡航控制器之系統設計與實現，國立中興大學電機工程學系所碩士論文。
[18] Kim, H. and Yi, K., "Design of a Model Reference Cruise Control Algorithm," SAE Int. J. Passeng. Cars Electron. Electr. Syst. 5(2):2012, doi:10.4271/2012-01-0492.
[19] Cantas, M.R., Gelbal, S.Y., Guvenc, L., and Aksun Guvenc, B., “Cooperative Adaptive Cruise Control Design and Implementation,” SAE Technical Paper 2019-01-0496, 2019, doi:10.4271/2019-01-0496
[20] 吳思蒨，2015，自走車於十字路口紅綠燈之影像辨識，國立成功大學工程科學系碩士論文。
[21] 江奇勳，2017，紅綠燈自動辨識在自走車之應用，國立成功大學工程科學系碩士論文。
[22] C. Yu, C. Huang and Y. Lang, "Traffic light detection during day and night conditions by a camera," IEEE 10th INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS, Beijing, 2010, pp. 821-824.
doi: 10.1109/ICOSP.2010.5655934
[23] TensorFlow from，https://www.tensorflow.org/
[24] TensorFlow llicens form https: //github.com/tensorflow/models/blob/ma-ster/LICENSE
[25] Deep learning- wiki from https://en.wikipedia.org/wiki/Deep_learning
[26] 斎藤康毅，Deep learning用Python進行深度學習的基礎理論實作，碁峰資訊股份有限公司，2018。
[27] D. H. Huble, T. N. Wiesel, "Receptive fields,binocular interaction and functional architecturein the cat's visual cortex," Journal of Physiol. Lond. 160, pp. 106-154,1962.
[28] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” in Advances in neural information processing systems, pp. 1097–1105, 2012.
[29] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, “Going deeper with convolutions,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1–9, 2015.
[30] Karen Simonyan, Andrew Zisserman” Very Deep Convolutional Networks For Large-Scale Image Recognition” ,Published as a conference paper at ICLR 2015
[31] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770–778, 2016
[32] Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy,
Scott Reed, Cheng-Yang Fu, Alexander C. Berg, “SSD: Single Shot MultiBox Detector ”European conference on computer vision. Springer, Cham, 2016.
[33] Howard A G,Zhu M,Chen B,et al.,Mobilenets:Efficient convolutional neural networks for mobile vision applications[J], arXiv preprint arXiv:1704.04861,2017.
[34] Tensorflow detection model zoo from: https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md
[35] Microsoft Common Objects in Context from http://cocodataset.org/#home
[36] PID controller from https://en.wikipedia.org/wiki/PID_controller
[37] PID 控制原理 from http://www.ni.com/zh-tw/innovations/white-papers/06/pidtheory-explained.html
[38] Pan, Sinno Jialin, and Qiang Yang. "A survey on transfer learning." IEEE.Transactions on knowledge and data engineering 22.10 (2010): 1345-1359.
[39] RGB color model from https://en.wikipedia.org/wiki/RGB_color_model
[40] HSL and HSV color model from https://en.wikipedia.org/wi-ki/HSL_and_HSV
[41] Perspective transformation from https://www.itread01.com/co-ntent/1542211447.html