高效能的道路線及前方車輛辨識能力是先進車輛輔助系統的核心技術，本研究基於連續影像的慣性特性，藉由圖像處理和分析技術，改善道路線與前車辨識效果，提高系統的強健性和穩定性，實現可靠且精確的道路線及前方車輛辨識演算法。
本論文提出的演算法分為道路線辨識、前方車輛辨識、連續訊框慣性追蹤三個部分。道路線辨識使用圖像處理和迴歸分析，搜尋道路線峰值並預測未知的道路線座標，同時利用道路與道路線顏色差異辨識道路線；前方車輛辨識運用特徵強化、特徵的連續性與XY軸大小，進行前車的分割和辨識；連續訊框慣性追蹤利用連續影像的相關性，提取前幀圖像中的道路線和前車特徵，應用於下幀圖像，標記搜尋到的相似道路線和前車。
分析各種行車影像的測試結果，本研究提出的演算法，可以減少額外的計算量，達到效能最佳化；連續影像的慣性追蹤，可以提高辨識的準確性、強健性，並有效降低執行時間。

High-performance identification of lane markings and forward vehicles is a critical technology for advanced driver assistance systems. This research is based on exploiting the consecutive images' inertial tracking characteristics and utilizes image processing and analysis techniques to enhance the effectiveness of lane marking and forward vehicles identification. The objective is to improve system robustness and stability, achieving a reliable and precise algorithm for identifying lane markings and forward vehicles.
The algorithm presented in this paper consists of three main components: lane marking identification, forward driver identification, and consecutive frame inertia tracking. For lane marking identification, image processing and regression analysis techniques are employed to detect lane marking peaks and predict unknown lane marking coordinates. Additionally, lane marking differentiation is achieved based on color contrast with the road surface. For forward driver identification, feature enhancement techniques, feature continuity, and XY-axis size are utilized for effective segmentation and identification of front vehicles. In the consecutive frame inertia tracking phase, correlations within consecutive images are leveraged to extract lane marking and forward driver features from preceding frames, which are then applied to search for similar road lanes and front vehicles in subsequent frames.
Through road image testing in various environments, the algorithm demonstrates improved performance by using single-channel lane marking identification and narrowed forward driver recognition, reducing additional computational burden. By employing consecutive image inertia tracking, the recognition accuracy and robustness are further improved, while the execution time is reduced.

第一章 緒論 1
1-1 前言 1
1-2 文獻回顧 2
1-3 動機與目的 5
1-4 研究方法 7
1-5 論文架構 9
第二章 道路線辨識 11
2-1 道路線偵測及預測 12
2-2 道路線顏色辨識 22
2-3 道路線預測座標修正及更新 26
2-3-1 道路線消逝點調整 29
第三章 前方車輛辨識 33
3-1 陰影檢測 34
3-1-1 左右車道陰影檢測 40
3-2 前車辨識 43
第四章 基於連續影像的道路線與前方車輛慣性追蹤 57
4-1 道路線慣性追蹤 57
4-2 前方車輛慣性追蹤 62
第五章 實驗及統計分析 67
5-1 模擬平台軟硬體規格及測試影片 67
5-1-1 測試平台軟硬體規格 67
5-1-2 演算法參數設定 68
5-1-3 行車測試影片 70
5-2 各種行車環境的測試 74
5-2-1 晴天 76
5-2-2 陰天 79
5-2-3 雨天 83
5-2-4 夜晚 87
5-3 實驗結果統計與分析 91
5-3-1 晴天 95
5-3-2 陰天 100
5-3-3 雨天 105
5-3-4 夜晚 109
5-4 實驗結果比較 115
5-4-1 道路線偵測實驗結果比較 115
5-4-2 前方車輛偵測實驗結果比較 117
第六章 結論與未來方向 119
6-1 結論 119
6-2 未來方向 120
參考文獻 121
作者簡歷 125

[1]J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 27-30 June 2016.
[2]C. Ning, H. Zhou, Y. Song, and J. Tang, “Inception Single Shot MultiBox Detector for object detection,” 2017 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), 10-14 July 2017.
[3]Y. Li, A. Dua and F. Ren, “Light-Weight RetinaNet for Object Detection on Edge Devices,” 2020 IEEE 6th World Forum on Internet of Things (WF-IoT), 02-16 June 2020.
[4]K. Duan, S. Bai, L. Xie, H. Qi, Q. Huang, and Q. Tian, “CenterNet: Keypoint Triplets for Object Detection,” 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 27 -02 October -November 2019.
[5]D. Qiao, X. Wu, and T. Wang, “A Lane Recognition Based on Line-CNN Network,” 2020 Asia-Pacific Conference on Image Processing, Electronics and Computers (IPEC), 14-16 April 2020.
[6]K. Rebai, N. Achour, and O. Azouaoui, “Hierarchical SVM classifier for road intersection detection and recognition,” 2013 IEEE Conference on Open Systems (ICOS), 02-04 December 2013.
[7]J. Zheng, S. Yang, X. Wang, X. Xia, Y. Xiao, and T. Li, “A Decision Tree Based Road Recognition Approach Using Roadside Fixed 3D LiDAR Sensors,” IEEE Access ( Volume: 7), pp. 53878 - 53890, 22 April 2019 .
[8] K. Zhao, Q. Han, C.-B. Zhang, J. Xu, and M.-M. Cheng, “Deep Hough Transform for Semantic Line Detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence ( Volume: 44, Issue: 9, 01 September 2022), pp. 4793 - 4806, 03 May 2021 .
[9]M. Marzougui, A. Alasiry, Y. Kortli, and J. Baili, “A Lane Tracking Method Based on Progressive Probabilistic Hough Transform,” IEEE Access ( Volume: 8), pp. 84893 - 84905, 04 May 2020 .
[10]M. H. Hasan and D. K. Das, “Analysis of Edge Detection for Road Lanes through Hardware Implementation,” 2021 Fourth International Conference on Electrical, Computer and Communication Technologies (ICECCT), 15-17 September 2021.
[11]紀富翔, 以電腦視覺為主的快速前車偵測與警示系統,國立東華大學碩士論文, 1 102.
[12]許智誠, 結合慣性道路線標記與車輛陰影之前車偵測演算法, 國立東華大學碩士論文, 7 100.
[13]H. Wang, Y. Wang, X. Zhao, G. Wang, H. Huang, and J. Zhang, “Lane Detection of Curving Road for Structural Highway With Straight-Curve Model on Vision,” IEEE Transactions on Vehicular Technology ( Volume: 68, Issue: 6, June 2019), pp. 5321 - 5330, 26 April 2019 .
[14]T.-Y. Sun and W.-C. Huang, “Embedded vehicle lane-marking tracking system,” 2009 IEEE 13th International Symposium on Consumer Electronics, 25-28 May 2009.
[15]L.-W. Wang, W.-C. Siu, X.-F. Yang, Z.-S. Liu, and D. P.-K. Lun, “Highly Reliable Vehicle Detection through CNN with Attention Mechanism,” 2022 IEEE International Conference on Consumer Electronics (ICCE), 07-09 January 2022.
[16]X. Li and X. Guo, “A HOG Feature and SVM Based Method for Forward Vehicle Detection with Single Camera,” 2013 5th International Conference on Intelligent Human-Machine Systems and Cybernetics, 26-27 August 2013.
[17]L.-W. Wang, X.-F. Yang, and W.-C. Siu, “Learning Approach with Random Forests on Vehicle Detection,” 2018 IEEE 23rd International Conference on Digital Signal Processing (DSP), 19-21 November 2018.
[18]Y. Munian, A. Martinez-Molina, and M. Alamaniotis, “Intelligent System for Detection of Wild Animals Using HOG and CNN in Automobile Applications,” 2020 11th International Conference on Information, Intelligence, Systems and Applications (IISA, 15-17 July 2020.
[19] A. D. X, A. N. Titus amd A. A, “Vehicle License Plate Localization based on Local Binary Pattern Features,” 2019 International Conference on Recent Advances in Energy-efficient Computing and Communication (ICRAECC), 07-08 March 2019.
[20]T. Wang, C. Xiu, and Y. Cheng, “Vehicle recognition based on saliency detection and color histogram,” The 27th Chinese Control and Decision Conference (2015 CCDC), 23-25 May 2015.
[21]F.-H. Chi, C.-L. Huo, Y.-H. Yu, and T.-Y. Sun, “An effective approach for forward vehicle detection with tracking,” The 1st IEEE Global Conference on Consumer Electronics 2012, 02-05 October 2012.
[22]鄭鈺傑, 基於智慧型手機以前車輛對稱性與被是分類為基礎的前車辨識演算法, 國立東華大學碩士論文, 7 104.
[23]張家維, 以DSP平台實現之前方車道與車輛偵測系統, 國立交通大學碩士論文http://asiair.asia.edu.tw/ir/bitstream/310904400/5792/1/1063.pdf. [存取日期: 6 6 2023]
[24]K. P. Sinaga and M.-S. Yang, “Unsupervised K-Means Clustering Algorithm,” IEEE Access ( Volume: 8), pp. 80716 - 80727, 20 April 2020 .
[25]H. Tizhoosh, “Fast fuzzy edge detection,” 2002 Annual Meeting of the North American Fuzzy Information Processing Society Proceedings. NAFIPS-FLINT 2002 (Cat. No. 02TH8622), 27-29 June 2002.
[26]"監理法規檢索系統" [線上] Available
:https://www.mvdis.gov.tw/webMvdisLaw/LawArticle.aspx?LawID=A0011027&ShowType=MdfLaw. [存取日期: 6 6 2023].
[27]F.-H. Chi, C.-L. Huo, Y.-H. Yu, and T.-Y. Sun, “Forward vehicle detection system based on lane-marking tracking with fuzzy adjustable vanishing point mechanism,” 2012 IEEE International Conference on Fuzzy Systems, 10-15 June 2012.
[28]"道路交通安全規則"[線上] Available:
https://law.moj.gov.tw/LawClass/LawSingleRela.aspx?media=print&PCODE=K0040013&FLNO=80&ty=L. [存取日期: 6 6 2023].
[29]顏妤樺, 基於智慧型手機的自適應性道路線偏移警示系統, 國立東華大學碩士論文, 7 103.