(3.238.96.184) 您好！臺灣時間：2021/05/12 23:56

### 詳目顯示:::

:

• 被引用:0
• 點閱:66
• 評分:
• 下載:0
• 書目收藏:0
 為改善傳統霍夫轉換效率，本論文提出基於分區加性原理之快速霍夫轉換。首先，為降低因分散的雜點而產生誤判，將影像切為若干個區域。接著對每個區域內的cell計算其特徵點位置及權重，並依角度資訊於特定角度區間轉換，以近似原本霍夫轉換結果。而轉換的同時進行排序，每個區域皆限制其排序數量，並用峰值展開window的方式剔除相近的直線。最後將各區域霍夫轉換的結果對應回全域霍夫轉換結果，對應的同時將排序的資料對應回原影像直線即完成偵測。提出之演算法平均處理速率為5486(edge/ms)。相較傳統霍夫轉換，可節省95%以上運算時間。
 In this thesis, based on Additive Region Segmentation a fast Hough Transform is proposed. To decrease the undesired detected line due to noise, the image is partitioned into several regions. Moreover, the feature point and the weighting of cell within each region are provided to transform into specific section by its angle information. The transform and sorting are performed simultaneously, and every region is limited with its sorting number, and the similar lines are removed with the method window extended by peak. Further, the result of region HT is mapping to Global HT to find the line. The experiment results show that the performance of the proposed algorithm is 5486 edges per millisecond. The computing time is saving about 95% comparing with Standard Hough Transform.
 中文摘要 iAbstract ii目錄 iii圖目錄 v表目錄 vii第一章 緒論 11.1 研究動機 11.2 研究目的 11.3 論文架構 1第二章 文獻探討 22.1 Hough Transform簡介 22.2 HT類型及應用 4第三章 研究方法 73.1 演算法架構 73.2 梯度及角度分類 93.3 區域分割 133.4 特徵點判定 143.5 特定區間內霍夫轉換 183.6 區域至全域霍夫轉換對應 213.7 直線偵測 23第四章 實驗結果與討論 254.1 複雜區域之影響 254.2 轉換區間之影響 264.3 Window大小之影響 274.4 排序數量應用 284.5 運算量分析 304.6 準確率分析 34第五章 結論與未來工作 355.1 結論 355.2 未來工作 35參考文獻 36
 [1] Duda, R. O. and P. E. Hart, “Use of the Hough Transformation to Detect Lines and Curves in Pictures,” Comm. ACM, vol. 15, pp. 11–15, Jan 1972[2] Hough, P.V.C. Method and means for recognizing complex patterns, U.S. Patent 3,069,654, Dec. 18, 1962[3] Dana H. Ballard, “Generalizing the Hough transform to detect arbitrary shapes,” Pattern Recoqnition vol. 13, no. 2, pp. 111-122 ,1981[4] Leandro A.F. Fernandes, Manuel M. Oliveira “Real-time line detection through an improved Hough transform voting scheme,” Pattern Recogn., 41 (1) 2008, pp. 299–314[5] R. K. Satzoda, S. Suchitra, and T. Srikanthan “Parallelizing the Hough Transform Computation,” IEEE Signal Processing Letters, vol. 15, 2008[6] Ravi Kumar Satzoda, Suchitra Sathyanarayana, Thambipillai Srikanthan “Hierarchical Additive Hough Transform for Lane Detection,” IEEE Embedded Systems Letters, vol. 2, no. 2, Jun 2010[7] Shengzhi Du, Barend Jacobus van Wyk, Chunling Tu, and Xinghui Zhang “An Improved Hough Transform Neighborhood Mapfor Straight Line Segments,” IEEE Transactions On Image Processing, vol. 19, no. 3, Mar 2010[8] Zezhong Xu, Bok-Suk Shin, and Reinhard Klette “ Accurate and Robust Line Segment Extraction Using Minimum Entropy With Hough Transform,” IEEE Transactions On Image Processing, vol. 24, no. 3, Mar 2015[9] Shengzhi Du, Chunling Tu, Barend Jacobus van Wyk, and Zengqiang Chen “Collinear Segment Detection Using HT Neighborhoods,” IEEE Transactions On Image Processing, vol. 20, no. 12, Dec 2011[10] Hart, P. E., “How the Hough Transform was Invented,” IEEE Signal Processing Magazine, vol.26, Issue 6, pp 18 – 22 Nov 2009[11] [Online]. Available: http://www.cpubenchmark.net/cpu_list.php
 國圖紙本論文
 推文當script無法執行時可按︰推文 網路書籤當script無法執行時可按︰網路書籤 推薦當script無法執行時可按︰推薦 評分當script無法執行時可按︰評分 引用網址當script無法執行時可按︰引用網址 轉寄當script無法執行時可按︰轉寄

 無相關論文

 無相關期刊

 無相關點閱論文

 簡易查詢 | 進階查詢 | 熱門排行 | 我的研究室