(3.238.118.78) 您好!臺灣時間:2021/04/15 23:27
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:陳品叡
研究生(外文):Pin-Ruei Chen
論文名稱:應用AdaBoost辨識軌跡以改善生成之電子地圖
論文名稱(外文):Applying AdaBoost to Identify Trace and Improve the Generation of Digital Map
指導教授:張嘉銘張嘉銘引用關係
指導教授(外文):Chia-Ming Chang
口試委員:張嘉銘
口試委員(外文):Chia-Ming Chang
口試日期:2013-07-04
學位類別:碩士
校院名稱:大同大學
系所名稱:資訊工程學系(所)
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2013
畢業學年度:101
語文別:中文
論文頁數:40
中文關鍵詞:數位地圖AdaBoostGPS
外文關鍵詞:Digital MapAdaBoostGPS
相關次數:
  • 被引用被引用:0
  • 點閱點閱:162
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:35
  • 收藏至我的研究室書目清單書目收藏:0
運用電子地圖與GPS 定位技術進行導航之功能已非常普及,因此我們需要一精準且能經常更新之電子地圖與導航系統相互配合,以達到正確導航的作用。在其他相關研究中,有使用高精準度GPS 接收器或大量的軌跡資料,經處理並據此結果建構地圖。但是,使用一般平價的GPS 接收器記錄軌跡資料以建置地圖或許是更好的方法。
可是平價GPS 接收器所記錄的軌跡資料,包含著許多雜訊或漂移。若直接使用這些資料建構電子地圖,將會得到一個非常不精確的地圖。在先前的研究中使用 NURBS 曲線以改善由GPS 軌跡所產生之電子地圖的精準度。但是當用於定位的衛星數有增減時,所產生的定位點軌跡紀錄雖說以NURBS 曲線增進了準確度,但是修正效果仍然有限。
在本論文中將這類因用於定位的衛星數增減所產生的漂移,歸類為Z 型及α 型漂移。採用以類似AdaBoost Haar-like 尋找特徵的方式,找出符合Z 型及α 型特徵的軌跡片段,修改其HDOP 值,接續先前的處理流程以定位點及其HDOP 值作為 NURBS 曲線的控制點及權重,生成一條NURBS 曲線。不同權重可決定每個定位點對於曲線的影響程度,以降低漂移定位點之影響力,使軌跡更符合實際道路位置。
經過降低符合Z 型及α 型漂移特徵的軌跡片段的影響,可使產生出的NURBS 曲線更為平緩。較平緩的NURBUS 曲線代表更少的漂移,也因此加快累積軌跡的收斂效果,也就是降低藉由累積軌跡以得到電子地圖所需的軌跡資料量。
The application of navigation using digital map and GPS receiver is very common today. Therefore, a precise digital map that can be updated frequently is needed to cooperate with navigator. In the related researches, high precise GPS receivers or a lot of GPS data are used to construct a digital map. But using GPS log data collected from the low cost GPS receiver to construct the digital map may be a better way.
The location information logged using low cost GPS receivers is contained with a lot of noise and drift. If the raw GPS data are used directly, the constructed map may also contain many errors. In previous researches, the NURBS curve is applied to improve the accuracy of digital map generated from these GPS Traces. When the number of satellites used for positioning increases or decreases, the generated map using these logged data can not remove all these effects. Even applying NURBS curve to improve it, the correction effect is still limited.
In this thesis, these drifts are classified and named as Z-type and α-type drift. Using a method that is similar to AdaBoost Haar-like features identification to find the parts containing the Z-type and α-type drift. These HDOP values are adjusted to change their weights affecting the generation of NURBS curve. Then continued from previous researches, the location and HDOP of these GPS data are used as the control points of NURBS curve and the weight of control points, respectively. The weight of point will determine the influence to NURBS curve, thereby reducing the influence of drift of GPS data.
After reducing the influence of trace fragments containing Z-type and α-type drift, can produce a smoother NURBS curves. A smoother NURBS curves means less drift, therefore, the convergence speed of accumulation is faster. In other words, the digital map can be obtained using less accumulated traces.
第1章 緒論 1
1.1 簡介 1
1.2 研究動機 1
1.3研究目的 2
1.4章節結構 2
第2章 相關研究 3
2.1 軌跡紀錄 3
2.2 DOP (Dilution of Precision) 3
2.3 AdaBoost演算法 4
2.3.1 Haar-like特徵擷取 5
2.3.2 AdaBoost學習演算法 6
2.4 B-Spline曲線 8
2.5 NURBS曲線 9
2.6 地圖建構 10
2.6.1 幾何模型與B-Spline 近似值 10
2.6.2 以高精準度感測器自動建構地圖 12
2.6.3 以GPS軌跡建構地圖 12
第3章 系統架構 15
3.1 系統概要 15
3.1.1先前研究 15
3.1.2新增部份 16
3.2對原始軌跡進行處理 17
3.2.1可用來作為特徵的參數 17
3.2.2計算角度特徵值 19
3.2.3正規化 18
3.2.4計算速度特徵值 21
3.3 AdaBoost學習演算法 23
3.3.1正負樣本 23
3.3.2 AdaBoost學習演算法 23
3.4特徵比對並調整DOP值 26
3.4.1特徵比對 26
3.4.2調整DOP值 27
3.5 NURBS調整軌跡 27
第4章 實驗介紹及分析 29
4.1 實驗平台及開發工具 29
4.2測試軌跡資料庫 29
4.3 實驗分析 29
4.3.1樣本分段 29
4.3.2比對結果 31
4.3.3誤判及誤判修正效果 32
4.3.4 Z型、α型修正效果 34
4.3.5 累計軌跡之修正結果比較 35
第5章 結論與未來展望 38
5.1 結論 38
5.2 未來展望 38
參考文獻 39
[1] W.-L. Lin, “Application of NURBS Curve to Improve the Digital Map Generated
from GPS Traces,” Thesis for Master of Science, Tatung University, 2009.
[2] Kearns, M.J., and Vazirani, U.V., “An Introduction to Computational
Learning Theory,” The MIT Press, August 1994.
[3] Viola, P., and Jones, M.J., “Robust Real-Time Face Detetion,”
International Journal of Computer Vision, pp.137-154, 2004.
[4] Lienhart, R., and Maydt, J.,“An extended set of Haar-like features for
rapid object detection,” International Conference on Image Processing, I-
900-903 vol.1,2002.
[5] H.-A. Chou and C.-M. Chang, 「應用GPS軌跡與衛星位置資訊以增進電子地圖精確度, 」
2009 國立成功大學數位生活科技-人本生活科技, 2009.
[6] D. Ben-Arieh, S. Chang, and G. Zhang, “Geometric modeling of highways using
GPS data and B-Spline approximation,” Master’s thesis, Department of
Industrial and Manufacturing Systems Engineering, Kansas State University,
2002.
[7] U. Noyer, J. Schomerus, H. H. Mosebach, J. Gacnik, C. Loper and K. Lemmer,
”Generating High Precision Maps for Advanced Guidance Support,” in Proc.
2008 IEEE Intelligent Vehicles Symposium, pp. 871-876, June 2008.
[8] R. Bruntrup, S. Edelkamp, S. Jabbar, and B. Scholz, “Incremental map
generation with GPS traces,” in Proc. of the 8th Int’l IEEE Conf. on
Intelligent Transportation Systems, pp. 574–579, Sept. 2005.
[9] Tao Guo, Iwamura K., and Koga, M., “Towards high accuracy road maps
generation from massive GPS Traces data,” in Pro. IGARSS 2007. IEEE
International Geoscience and Remote Sensing Symposium, pp. 667-670, July
2007.
[10] C. Chen and Y. Cheng, “Roads Digital Map Generation with Multi-track GPS
Data,” conf. On 2008 Education Technology and Training, and International
Workshop on Geoscience and Remote Sensing, pp. 508-511, Dec. 2008.
[11] J. J. Davies , A. R. Beresford, and A. Hopper, “Scalable, Distributed,
Real-Time Map Generation,” IEEE Pervasive Computing, vol. 5, pp. 47-54,
Oct. 2006.
[12] B. Niehoefer, R. Burda, C. Wietfeld, F. Bauer, and O. Lueert, “GPS
Community Map Generation for Enhanced Routing Methods based on Trace-
Collection by Mobile Phones,” in Proc. of 2009 1st. Intl IEEE Conf. on
Advances in Satellite and Space Communications, SPACOMM, pp.156-161, July
2009.
[13] NMEA data.http://www.gpsinformation.org/dale/nmea.htm
[14] KML documentation. http://code.google.com/apis/kml/documentation/
[15] Google Earth. http://earth.google.com/
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔