系統識別號 | U0002-0308200915052600 |
---|---|
DOI | 10.6846/TKU.2009.00088 |
論文名稱(中文) | 強健性區域特徵應用於物件辨識 |
論文名稱(英文) | Efficient Wavelet-Based Scale Invariant Features for Object Recognition |
第三語言論文名稱 | |
校院名稱 | 淡江大學 |
系所名稱(中文) | 資訊工程學系碩士班 |
系所名稱(英文) | Department of Computer Science and Information Engineering |
外國學位學校名稱 | |
外國學位學院名稱 | |
外國學位研究所名稱 | |
學年度 | 97 |
學期 | 2 |
出版年 | 98 |
研究生(中文) | 林楠傑 |
研究生(英文) | Nan-Chieh Lin |
學號 | 696410777 |
學位類別 | 碩士 |
語言別 | 繁體中文 |
第二語言別 | 英文 |
口試日期 | 2009-06-22 |
論文頁數 | 47頁 |
口試委員 |
指導教授
-
顏淑惠(shyen@cs.tku.edu.tw)
委員 - 顏淑惠(shyen@cs.tku.edu.tw) 委員 - 林慧珍(hjlin@cs.tku.edu.tw) 委員 - 施國琛(tshih@cs.tku.edu.tw) |
關鍵字(中) |
比對 離散小波轉換 主要比例 比例不變性 極座標轉換 特徵點描述子 |
關鍵字(英) |
matching discrete wavelet transform (DWT) dominate scale (DS) scale invariance log-polar transform feature point descriptor |
第三語言關鍵字 | |
學科別分類 | |
中文摘要 |
物件辨識的方式有很多,其中常被人使用的方式為偵測物件特徵點然後對特徵點進行比對。然而影像常常會有旋轉、縮放或是平移的變動,因此在偵測特徵點的同時會受到這些變動的影響。本研究的目的在於找尋物件或是影像的特徵點及其特徵向量並且具有強健性,在影像經過變動之後仍然具有不變性。 本研究利用DWT找出特徵點,接著使用log-polar轉換使特徵具有角度的不變性,利用亮度的差值決定特徵向量的內容以抵抗亮度的改變,最後利用幾何學相似三角形的原理提升比對的正確率。 在實驗中與CCH[1]做比較,確實提昇了Scaling不變性的效果,另外對於亮度變化以及模糊化也有不錯的表現,此外其他的實驗和CCH[1]有著類似的結果。在時間上,跟CCH相差甚少,也就是說相較於SITF[3]快了近兩倍之多。 |
英文摘要 |
Feature points’ matching is a popular method in dealing with object recognition problems. However, variations of images, such as shift, rotation, and scaling, influence the matching correctness. Therefore, a feature point matching system with distinctive and invariant feature point detector as well as robust description mechanism becomes the main challenge of this issue. We use discrete wavelet transform (DWT) and accumulated map to detect feature points which are local maximum points on the accumulated map. DWT calculation is very efficient comparing to that of Harris corner detection or Difference of Gaussian (DoG) proposed by Lowe. Besides, feature points detected by DWT are located more evenly on texture area unlike those detected by Harris’ are clustered on corners. To be scale invariant, the dominate scale (DS) is determined for each feature point. According to the DS of a feature point, an appropriate size of region centered at this feature point is transformed to log-polar coordinate system to improve the rotation and scale invariance. A descriptor of dimension 32 is made of the contrast information to enhance the illumination robustness. Finally, in matching stage, a geometry relation is adopted to improve the matching accuracy. Comparing to existing methods, the proposed algorithm has better performance especially in scale invariance and robustness to blurring effect. |
第三語言摘要 | |
論文目次 |
目錄 淡江大學論文中文提要..........................I 淡江大學論文英文提要.........................II 目錄........................................III 圖目錄........................................V 表目錄......................................VII 第一章 緒論...................................1 1.1 研究動機與目的............................1 1.2 研究內容..................................2 1.3 論文架構..................................3 第二章 相關研究...............................4 2.1 擷取特徵點................................4 2.1.1 Harris Corner Detector..................4 2.1.2 Scale Invariant Feature Transformation..5 2.1.3 利用DTCWT找尋特徵點.....................7 2.2 Log-Polar轉換. ............................8 第三章 提出的方法............................11 3.1 偵測特徵點...............................11 3.2 計算特徵向量.............................15 3.2 比對特徵向量.............................18 第四章 實驗結果與討論........................21 第五章 結論與未來展望........................30 參考文獻.....................................31 附錄 英文論文................................32 圖目錄 圖1.1實驗流程圖...............................3 圖2.1 Moravec Corner Detector 示意圖..........5 圖2.2 DoG示意圖...............................6 圖2.3 SIFT臨點示意圖..........................6 圖2.4 Fauqueur所提出特徵點偵測的方法..........8 圖2.5 Accumulated Map示意圖...................8 圖2.6 Log-polar轉換...........................9 圖2.7 Log-polar旋轉產生的位移.................9 圖3.1 輸入影像進行DWT示意圖..................12 圖3.2 DS示意圖...............................14 圖3.3 找特徵點流程示意圖.....................15 圖3.4 計算特徵向量示意圖.....................17 圖 4.1 實驗dataset...........................22 圖4.2 影像變化dataset........................22 圖4.3 Scaling recall&precision 比較圖........23 圖4.4 模糊化 recall&precision 比較圖.........24 圖4.5 提出的方法與Harris Detector比較圖......25 圖4.6亮度變化recall&precision 折線圖.........26 圖4.7旋轉180° recall&precision 比較圖........26 圖4.8旋轉5° recall&precision 比較圖..........27 圖4.9 加入高斯雜訊 recall&precision 比較圖...27 圖4.10 JPEG recall&precision 比較圖..........28 表目錄 表4.1 計算速率之比較表.......................29 |
參考文獻 |
[1] H. P. Moravec. Towards Automatic Visual Obstacle Avoidance. Proc. 5th International Joint Conference on Artificial Intelligence, pp. 584, 1977. [2] H. P. Moravec. Visual Mapping by a Robot Rover. International Joint Conference on Artificial Intelligence, pp. 598-600, 1979. [3] Lowe, David G. (1999). "Object recognition from local scale-invariant features". Proceedings of the International Conference on Computer Vision 2: 1150–1157. [4] Mikolajczyk, K.; Schmid, C. (2005). "A performance evaluation of local descriptors". IEEE Transactions on Pattern Analysis and Machine Intelligence 27: 1615–1630. [5] N G Kingsbury, “Complex wavelets for shift invariant analysis and filtering of signals,” Journal of Applied and Computational Harmonic Analysis, vol 10, no 3, May 2001, pp. 234-253. [6] E. Loupias, N.Sebe,.S. Bres, J.-M.Jolion, “Wavelet-based salient points for image retrieval,” Int. Conf. on Image Processing, 2000, pp. 518-521. [7] Chun-Rong Huang, Chu-Song Chen and Pau-Choo Chung, “Contrast context histogram---An efficient discriminating local descriptor for object recognition and image matching”, Pattern Recognition 41 (2008), pp. 3071 – 3077. [8] Chris Harris and Mike Stephens, “A combined corner and edge detector”, In Proceedings of The Fourth Alvey Vision Conference (1988), pp. 147-151. [9] David G. Lowe, “Distinctive image features from scale-invariant keypoints”, International Journal of Computer Vision Volume 60, Issue 2 (2004), pp. 91 - 110. [10] Julien Fauqueur, Nick Kingsbury and Ryan Anderson, “Multiscale keypoint detection using the dual-tree complexwavelet transform”, IEEE Conference on Image Processing, Atlanta, GA, 8-11 Oct(2006). [11] Chi-Man Pun and Moon-Chuen Lee, “Log-polar wavelet energy signatures for rotation and scale invariant texture classification”, Pattern Analysis and Machine Intelligence, IEEE Transactions Volume: 25, Issue 5(2003), pp. 590- 603. |
論文全文使用權限 |
如有問題,歡迎洽詢!
圖書館數位資訊組 (02)2621-5656 轉 2487 或 來信