Real-time Long-range Lane Detection and Tracking for Intelligent Vehicle
===
###### Xin Liu, Bin Dai, Jinze Song, Hangen He The Institute of Automation National University of Defense Technology Changsha, China [e-mail]

## Keyword
lane detection; lane tracking; intelligent vehicle; lane marking segmentation

## Content
本文提出了一種實時的遠程車道檢測和跟踪方法，以滿足在高速公路上行駛的高速智能車的需求。
基於線性拋物線兩車道公路模型和名為車道標記分割的新型強車道標記功能，該方法的最大車道檢測距離可達120米。
通過使用卡爾曼濾波器估算自我車輛的橫向偏移量來選擇和跟踪車道線。
從高速公路實際交通場景中提取測試數據集的實驗結果表明，本文提出的方法能夠以較低的時間成本實現較高的檢測率。

### 實驗方法
#### 1. 道路偵測
1. 使用顏色、邊緣等方式二值化取出影像中白色的物件。
2. 將符合車道寬度的白色物件列為候選車道線

3. 藉由車道的線性關係將車道連接
4. 套用遠近車道線模型確認哪些是確定的車道線
$u_{x}=\left\{\begin{array}{cc}\alpha_{x}+\beta_{x}\left(v_{x}-v_{n}\right), & \text { if } \quad v_{x} \geq v_{n} \\ \alpha_{x}+\beta_{x}\left(v_{x}-v_{n}\right)+\gamma_{x}\left(v_{x}-v_{n}\right)^{2}, & \text { if } v_{f} \leq v_{x} \leq v_{n}\end{array}\right.$
$v_{x}:垂直座標 \\
\alpha,\beta:線性參數\\
\gamma:曲率$
#### 2. 車道追蹤
1. 使用車道的飽和度分類出分斷線與邊線，定且定義分段線為中間線
$L_{x} \in\left\{\begin{array}{ll}S L, & \text { if } f\left(L_{x}\right)>0.6 \\ S G, & \text { if } f\left(L_{x}\right) \leq 0.6\end{array}\right.$
2. 確認車道線後，以畫面中心做為汽車位置點並藉由Kalman filter預測出物體的位置的坐標及速度，最後依據二車道線與車子位置的變化量作為偏移量。

### 結果
#### 實驗環境
適用於高速公路，並且只取鄰近車子的左、右車道作為判斷依據
#### 實驗結果
FS:總測試幀數
LS:實際車道數
DL:找到的車道數
FP:(假陽性)判斷錯誤的數量
### 結論
本文提出實時遠程車道檢測基於線性拋物線雙車道公路道路模型的方法，實驗結果表明，提出的方法可以達到高檢測率，極低的假陽性率和低時間成本。

## Review
本篇主要為偵測距離長的車道線，以往的研究都集中在距離不超過60m附近的車道，本文提出基於線性拋物線兩車道公路模型及新穎的車道標記功能檢測方法，將符合車道線的車道標記出來，再藉由車道線性關係將候選車道連接再一起，最後套用車道模型確定車道線。

## Reference
[1] W. Kwon and S. Lee, “Performance evaluation of decision making strategies for an embedded lane departure warning system,” Robot. Syst., vol. 19, no. 10, pp. 499–509, Sep. 2002.
[2] J. C. McCall and M. M. Trivedi, “Video-based lane estimation and tracking for driver assistance: Survey, system, and evaluation,” IEEE Trans. Intell. Transp. Syst., vol. 7, no. 1, pp. 20–37, Mar. 2006.
[3] J. McCall and M. M. Trivedi, “Visual context capture and analysis for driver attention monitoring,” in Proc. IEEE Conf. Intell. Transp. Syst., Washington, DC, Oct. 2004, pp. 332–337.
[4] D. D. Salvucci, “Inferring driver intent: A case study in lane-change detection,” in Proc. Human Factors Ergonomics Society 48th Annu. Meeting, New Orleans, LA, 2004, pp. 2228–2231.
[5] M. Bertozzi and A. Broggi, “GOLD: A parallel real-time stereo vision system for generic obstacle and lane detection,” IEEE Trans. Image Process., vol. 7, no. 1, pp. 62–81, Jan. 1998.
[6] A. Amditis, et. Al., “A holistic approach to the integration of safety applications: The INSAFES subproject within the European Framework Programme 6 integrating project PReVENT,” IEEE Trans. Intell. Transp. Syst., vol. 11, no. 3, pp. 554–566, Sep. 2010.
[7] A. Eidehall and F. Gustafsson, “Obtaining reference road geometry parameters from recorded sensor data,” in Proc. IEEE Intell. Veh. Symp., 2006, pp. 256–260.
[8] Y. Wang, E. K. Teoh, and D. Shen, “Lane detection and tracking using B-Snake,” Image Vis. Comput., vol. 22, no. 4, pp. 269–280, Apr. 2004.
[9] Q. Chen and H. Wang, “A real-time lane detection algorithm based on a hyperbola-pair model,” in Proc. IEEE Intell. Veh. Symp., 2006, pp. 510–515.
[10] W. Yan, B. Li and F. Michael, “Robust road modeling and tracking using condensation,” IEEE Trans. Intell. Transp. Syst., vol. 9, no. 4, pp. 570–579, Dec. 2008.
[11] W. Hendrik, L. Philipp and W. Gerd, “Vehicle tracking with lane assignment by camera and lidar sensor fusion,” in Proc. IEEE Intell. Veh. Symp., 2009, pp. 513–520.
[12] C. R. Jung and C. R. Kelber, “A robust linear-parabolic model for lane following.” in Proc. SIBGRAPI, 2004, pp. 72–79.
[13] N. Simond, “Reconstruction of the road plane with an embedded stereorig in urban environments,” in Proc. IEEE Intell. Veh. Symp., 2006, pp. 70–75.
[14] Y. He, H. Wang, and B. Zhang, “Color-based road detection in urban traffic scenes,” IEEE Trans. Intell. Transp. Syst., vol. 5, no. 4, pp. 309–318, Dec. 2004.
[15] S. S. Blackman, “Multiple-target tracking with radar applications,” Artech House, Norwood, MA, 1986..

###### tags: `Finish` `論文` `IEEE` `Conference`