現今日常生活中，行車駕駛者時常因為不熟悉道路的路面狀況，或是行車時駕駛者視線範圍縮小等原因，造成駕駛者無法確切得知路面狀況以有效閃避顛簸路段，導致行車舒適度下降。另外，路面顛簸所導致的交通事故時有所聞，因此道路品質的養護以及行車安全的提升儼然成為人們關注的議題。
本論文將以高爾夫球車為實驗平台，搭載網路攝影機、雷射測距儀、慣性量測儀與衛星定位儀來建構一套智慧型道路檢測系統。本系統主要分為三大功能(一)地形分類(二)坑洞偵測(三)路面品質分析。在地形分類方面，搭載網路攝影機來收集前方影像資訊，將利用此資訊為倒傳遞類神經網路(Back Propagation Neural Network, BPNN)之輸入來做為地形分類的訓練以及最終的分類機制。在坑洞偵測與路面品質分析方面，搭載雷射測距儀與網路攝影機以及慣性量測儀用以判斷路面坑洞與分析路面品質。系統最終將以衛星定位儀之經緯度資訊將此三大功能之結果標示於人機介面之Google Map上，可讓駕駛者得知附近路段的路況，並選擇最佳的路線來行駛。
本論文為第一套結合地形分類、坑洞偵測與路面品質分析之智慧型道路檢測系統，此系統讓駕駛者可以提前得知前方路面資訊，並避開路面不佳之路段，以降低交通事故的發生，也可將此資訊提供給政府相關單位，以作為是否需路面養護的依據，讓駕駛者有更安全舒適的行車環境。

Nowadays, the drivers cannot dodge bumpy road because of unfamiliar with traffic and poor visibility and to cause the traffic accident, therefore the situation of the roadway and driving safety are the most interested topic.
In this thesis, the main experimental equipment is golf car which equip the webcam, laser range finder, IMU and RTK-DGPS to construct an intelligent roadway detection system. In this system, we divide it into three functions, terrain classification, pothole detection and roadway quality analysis. In terms of terrain classification, the experimental equipment captures the front of image through the webcam, and this information as the inputs of Back Propagation Neural Network (BPNN) is the training of the terrain classification and the final classification mechanism. In terms of pothole detection and roadway quality analysis, the experimental equipment gauges the pothole and analyze the roadway quality through laser range finder, webcam and IMU. At the end, the system will gather the outcome of functions and then mark on the latitude and longitude of Google Map through RTK-DGPS on user interface to notify the drivers of nearby traffic.
This thesis, intelligent roadway detection system, is the first system which integrates the terrain classification, pothole detection and roadway quality analysis. This system provides the information of the front of roadway for drivers to avoid the rough roadway and to decrease the chance of the traffic accident, and the more safety and comfortable driving environment.

論文審定書 i
致 謝 ii
摘 要 iii
Abstract iv
圖 次 viii
表 次 xi
第一章 緒 論 1
1-1 研究動機 1
1-2 文獻回顧 2
1-3 主要貢獻 9
1-4 章節介紹 10
第二章 系統架構 11
2-1 實驗平台 11
2-2 感測器 12
2-2-1 網路攝影機(Webcam) 12
2-2-2 雷射測距儀（Laser Range Finder，LRF） 12
2-2-3 慣性量測儀(Inertial Measurement Unit，IMU) 15
2-2-4 全球衛星定位系統（Global Positioning System , GPS） 17
第三章 系統設計 19
3-1 系統概述 19
3-2 地形分類系統 21
3-2-1 特徵萃取 22
3-2-2 倒傳遞類神經網路 27
3-3 坑洞偵測與路面品質分析系統 29
A. 坑洞偵測系統 30
3-3-1 以網路攝影機影像演算法辨識路面坑洞 31
3-3-2 以雷射測距儀距離資訊辨識路面坑洞 32
3-3-3 以慣性量測儀加速度資訊辨識路面坑洞 33
3-3-4 以三感測器資訊融合於辨識坑洞 34
B. 路面品質分析系統 36
3-3-5 以慣性量測儀分析路面品質 38
3-3-6 以雷射測距儀分析路面品質 38
3-3-7 以多感測器融合分析路面品質 39
3-3-8 以衛星定位儀將坑洞與路面品質結果標示於Google Map上 40
第四章 系統實現 42
4-1 地形分類系統 42
4-1-1 四種地形分類 42
4-1-2 五種地形分類 43
4-1-3 六種地形分類 44
4-1-4 八種地形任取五種地形分類 45
4-1-5 八種地形分類 46
4-1-6 分類結果討論 48
4-2 坑洞偵測與路面品質分析系統 48
A. 坑洞偵測系統 48
4-2-1 平地路面 48
4-2-2 上下坡路面 50
4-2-3 雨天路面 51
B. 路面品質分析系統 53
4-2-4路面品質分析結果 54
4-3 人機介面 60
第五章 結論與未來展望 62
5-1 結論 62
5-2 未來展望 62
參考文獻 64

[1] 10大坑洞路 同列易肇事路段
https://tw.news.yahoo.com/10大坑洞路-同列易肇事路段-202345657.html
[2] A. Anelia, L. Matthies, D. Helmick, and P. Perona, “Fast terrain classification using variable-length representation for autonomous navigation,” IEEE Conference on Computer Vision and Pattern Recognition, pp. 1-8, 2007.
[3] L. Lu, C. Ordonez, E. G. Collins, E. Coyle, and D. Palejiya, “Terrain surface classification with a control mode update rule using a 2D laser stripe-based structured light sensor,” Robotics and Autonomous Systems, pp. 954-965, 2011.
[4] E. G. Collins, and E. J. Coyle, “Vibration-based terrain classification using surface profile input frequency responses,” IEEE International Conference on Robotics and Automation, pp. 3276-3283, 2008.
[5] S. M. Abbas, A. Muhammad, S. A. Mehdi, and K. Berns, “Improvements in accuracy of single camera terrain classification,” IEEE International Conference on Advanced Robotics, pp. 1-6, 2013.
[6] L. Bin, L. Yibin, and R. Xuewen, “The visual terrain classification algorithm based on fast neural networks and its application,” IEEE Chinese Control Conference, pp. 5780-5784, 2013.
[7] P. Giguere and G. Dudek, “A simple tactile probe for surface identification by mobile robots,” IEEE Transactions on Robotics, pp. 534-544, 2011.
[8] H. Oliveira and P. L. Correia, “Automatic road crack detection and characterization,” IEEE Transactions on Intelligent Transportation Systems, pp.155-168, 2013.
[9] C. Koch and I. Brilakis, “Pothole detection in asphalt pavement images,” Advanced Engineering Informatics, pp. 507-515, 2011.
[10] Y. H. Tseng, S. C. Kang, J. R. Chang, and C. H. Lee, “Strategies for autonomous robots to inspect pavement distresses,” Automation in Construction, pp. 1156-1172, 2011.
[11] X. Yu and E. Salari, “Pavement pothole detection and severity measurement using laser imaging,” IEEE International Conference on Electro/Information Technology, pp. 1-5, 2011.
[12] M. Fazeen, B. Gozick, R. Dantu, M. Bhukhiya, and M. C. González, “Safe driving using mobile phones,” IEEE Transactions on Intelligent Transportation Systems, pp. 1462-1468, 2012.
[13] P. Mohan, V. N. Padmanabhan, and R. Ramjee, “Nericell: rich monitoring of road and traffic conditions using mobile smartphones,” ACM conference on Embedded network sensor systems, pp. 323-336, 2008.
[14] V. Astarita, M. V. Caruso, G. Danieli, D. C. Festa, V. P. Giofrè, T. Iuele, and R. Vaiana, “A mobile application for road surface quality control: UNIquALroad,” Procedia-Social and Behavioral Sciences, pp. 1135-1144, 2012.
[15] A. Cigada, F. Mancosu, S. Manzoni, and E. Zappa, “Laser-triangulation device for in-line measurement of road texture at medium and high speed,” Mechanical Systems and Signal Processing, pp. 2225-2234, 2010.
[16] J. Eriksson, L. Girod, B. Hull, R. Newton, S. Madden, and H. Balakrishnan, “The pothole patrol: using a mobile sensor network for road surface monitoring,” ACM International Conference on Mobile Systems, pp. 29-39, 2008.
[17] F. Orhan and P. E. Eren, “Road hazard detection and sharing with multimodal sensor analysis on smartphones,” IEEE International Conference on Next Generation Mobile Apps, Services and Technologies, pp. 56-61, 2013.
[18] Club Car : http://www.clubcar.com/us/en/home.html.
[19] Camera : http://www.logitech.com/zh-tw/product/hd-pro-webcam-c920
[20] SICK : http://www.orbit.com.tw/download/LMS-291_Spec_Cht_02.pdf.
[21] IMU : http://www.xsens.com
[22] DGPS原理 : http://gnss.cnce.ntou.edu.tw/class8
[23] GPS :http://www.sokkia.com.tw/pd/gps/NovAtel/OEM/PROPAK-G2/G2 LBPLUSp1.htm.
[24] HSV : http://en.wikipedia.org/wiki/HSL_and_HSV
[25] T. Ojala, M. Pietikainen, and T. Maenpaa, “Multiresolution gray-scale and rotation invariant texture classification with local binary patterns,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 971-987, 2002.
[26] Z. J. Xu, M. Song, S. L. Li, and F. C. Sun, “Combining features for adaptive terrain classification based on ART neural network,” IEEE Chinese Control Conference, pp. 4808-4813, 2012.
[27] Y. N. Khan, P. Komma, and A. Zell, “High resolution visual terrain classification for outdoor robots,” IEEE International Conference on Computer Vision Workshop, pp. 1014-1021, 2011.
[28] 王進德，類神經網路與模糊控制理論入門與應用，全華出版社，2011。
[29] S. Theodoridis, A. Pikrakis, K. Koutroumbas, and D. Cavouras, “Introduction to pattern recognition : a MATLAB approach,” Academic Press, 2009.
[30] S. S. Haykin, “Neural networks : a comprehensive foundation,” Prentice Hall International, 1999.
[31] L. Wei, C. Cappelle, and Y. Ruichek, “Unscented information filter based multi-sensor data fusion using stereo camera, laser range finder and GPS receiver for vehicle localization,” IEEE Conference on Intelligent Transportation Systems, pp. 1923-1928, 2011.
[32] P. Aksamit and M. Szmechta, “Distributed, mobile, social system for road surface,” IEEE International Symposium on Computational Intelligence and Intelligent ]Informatics, pp. 37-40, 2011.