(3.236.118.225) 您好!臺灣時間:2021/05/16 11:03
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:陳啟禎
研究生(外文):Chii-Jen Chen
論文名稱:利用分水嶺轉換和區域鄰接圖的區域基礎視訊搜尋
論文名稱(外文):Region-based Video Retrieval Using Watershed Transformation and Region Adjacency Graph
指導教授:張瑞峰張瑞峰引用關係
指導教授(外文):Ruey-Feng Chang
學位類別:碩士
校院名稱:國立中正大學
系所名稱:資訊工程研究所
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:2001
畢業學年度:89
語文別:英文
論文頁數:47
中文關鍵詞:影像搜尋影像分割視訊搜尋分水嶺轉換區域鄰接圖圖形比對場景轉換偵測
外文關鍵詞:image retrievalimage segmentationvideo retrievalwatershed transformationregion adjacency graphgraph matchingscene change detection
相關次數:
  • 被引用被引用:0
  • 點閱點閱:194
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
  在這篇論文中,我們採用以區域為基礎的影像搜尋的概念應用於視訊的領域裡,並且提出了一個有效的以區域為基礎的視訊搜尋方法。所謂以區域為基礎的視訊搜尋,其目的是利用已經做好分割區域動作的一連串影像串列,並根據這些區域的特徵外貌及鄰近關係,來找出與待尋之影像物件具有最大相似數量的畫面。這項方法的主要構想是結合了分水嶺轉換和區域鄰接圖的方法。一開始,在影像串列中的每一張畫面,都先用分水嶺轉換的方法將畫面分割成數個有意義的區域。在做完分水嶺轉換的動作後,區域間的空間關係可以用區域鄰接圖來表示。在區域鄰接圖中,每個點代表一個區域,而每一條邊就代表相鄰區域的關係。在執行搜尋的過程中,我們引用了一種有效的圖論配對的演算法來比較待尋之影像與每一張畫面所對應的區域鄰接圖。
  為了增快物件的搜尋時間,我們也採用一種新的畫面埸景轉換偵測的技術。這個技術是利用每張畫面所對應的區域鄰接圖來做圖論配對的比較,並決定何時該發生場景轉換。另外為增加場景轉換偵測方法的準確度,根據兩個相鄰的區域鄰近圖之點與邊的新增和減少,來產生兩個小的子區域鄰近圖。在此兩個小的子區域鄰接圖,稱作『新加入的子區域鄰接圖』及『已消失的子區域鄰接圖』。最後相似性的評估取決於待尋影像的區域鄰近圖與影像區間中之關鍵影像的距離分數及上述兩子區域鄰接圖的距離分數之總合。我們可以從不同的影片測試檔案的實驗結果中,發現我們所提出的這個搜尋方法是相當有效的。

  In this paper we adopt the idea of region-based image retrieval into the video domain and propose an effective region-based video retrieval method. The goal of region-based video retrieval is to find the frames in an image sequence containing regions that match with the regions of the query image object having maximal similarity in region features and region adjacency relations. The main idea of our method is to combine the watershed transformation and the region adjacency graph (RAG). Each frame in an image sequence is first divided into several meaningful regions by the watershed transformation. After the watershed transformation, these regions and their spatial relationships of each frame will be represented with an RAG. In an RAG, a node represents a region and the spatial relationship of two neighboring regions is represented by an edge. In the retrieval processing, we adopt an efficient graph-matching algorithm to compare the RAG of a query image with the RAG of each frame.
  In order to improve the query time for the video retrieval, the scene change detection is adopted in the proposed method. In the new scene change detection method, the RAGs for neighboring frames are compared to decide whether a scene change occurs. To increase the accuracy rate for the scene change detection, two new smaller subRAGs, addition and deletion subRAGs depending on the addition and the deletion for the nodes and edges, can be generated from two neighboring RAGs. The evaluation of the similarity between two RAGs is to combine the distance score between the query image and key frame and these two subRAG’s distance scores. It is quite effective for our proposed method in experimental results on several test video sequences.

摘 要...................................................i
ABSTRACT....................................................iii
ACKNOWLEDGEMENTS..............................................v
TABLE OF CONTENTS............................................vi
LIST OF FIGURES.............................................vii
LIST OF TABLES................................................x
CHAPTER 1INTRODUCTION.........................................1
CHAPTER 2
WATERSHED TRANSFORMATION AND REGION ADJACENCY GRAPH...........5
 2.1 Watershed Transformation.............................5
 2.2 Over-Segmentation Problem............................9
 2.3 Region Adjacency Graph..............................10
 2.4 Graph-Matching Method for RAGs......................12
CHAPTER 3
THE PROPOSED REGION-BASED VIDEO RETRIEVAL METHOD.............18
 3.1 Color Watershed Transformation......................18
 3.2 The Proposed Video Retrieval........................22
 3.3 Scene Change Detection..............................23
 3.4 Different Subgraphs for RAGs........................27
CHAPTER 4EXPERIMENTAL RESULTS................................30
CHAPTER 5CONCLUSION..........................................43
REFERENCES...................................................45

[1]W. Niblack et al., “The QBIC project: Querying images by content using colour, texture and shape,” in Proceedings of the SPIE on Storage and Retrieval for Image and Video Databases, 1993, pp. 173-187.
[2]S. Ardizzoni, I. Bartolini and M. Patella, “Windsurf: Region-based image retrieval using wavelets,” in Proceedings of the IEEE 10th International Workshop on Database and Expert Systems Applications, Florence, Italy, Sept. 1999, pp. 167-173.
[3]P. Salembier and F. Marques, “Region-based representations of image and video: Segmentation tools for multimedia services,” IEEE Trans. Circuits Syst. for Video Technology, vol. 9, no. 8, pp. 1147-1169, Dec. 1999.
[4]M. K. Shan and S. Y. Lee, “Content-based video retrieval based on similarity of frame sequence,” in Proceedings International Workshop on Multimedia Database Management Systems, Dayton, Ohio, Aug. 1998, pp. 90-97.
[5]Y. Hiwatari, K. Fushikida and H. Waki, “An index structure for content-based retrieval from a video database,” in Proceedings on Computational Intelligence and Multimedia Applications, 1999, pp. 268-272.
[6]Y. A. Aslandogan and C. T. Yu, “Techniques and systems for image and video retrieval,” IEEE Trans. Know. Data Engineering, vol. 11, no. 1, pp. 56-63, Jan. 1999.
[7]E. K. Kang, S. J. Kim and J. S. Choi, “Video retrieval based on scene change detection in compressed streams,” IEEE Trans. Consumer Electronics, vol. 45, no. 3, pp. 932-936, Aug. 1999.
[8]D. Wang, “Unsupervised video segmentation based on watersheds and temporal tracking,” IEEE Trans. Circuits Syst. Video Technol., vol. 8, no. 5, pp. 539-546, Sept. 1998.
[9]L. Vincent and P. Soille, “Watersheds in digital spaces: An efficient algorithm based on immersion simulations,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 13, no. 6, pp. 583-598, June 1991.
[10]M. Baccar, L. A. Gee, R. C. Conzalez, and M. A. Abidi, “Segmentation of range images via data fusion and morphological watersheds,” Pattern Recognition, vol. 29, no.10, pp. 1673-1687, Feb.1996.
[11]D. Wang, “A multiscale gradient algorithm for image segmentation using watersheds,” Pattern Recognition, vol. 678, no. 12, pp. 2043—2052, Jan. 1997.
[12]L. Shafarenko, M. Petrou and, J. Kittler, “Automatic watershed segmentation of randomly textured color images,” IEEE Trans. Image Processing, vol. 6, no. 11, pp. 1530-1544, Nov. 1997.
[13]K. Haris, S. N. Efstratiadis, and N. Maglaveras, “Watershed-based image segmentation with fast region merging,” in Proceedings of the IEEE International Conference on Image Processing, Chicago, Illinois, Oct. 1998, vol. 3, pp. 338-342.
[14]K. Haris, S. N. Efstratiadis, N. Maglaveras, and A. K. Katsaggelos, “Hybrid image segmentation using watersheds and fast region merging,” IEEE Trans. Image Processing, vol. 7, no. 12, pp. 1684-1699, Dec. 1998.
[15]J. M. Gauch, “Image segmentation and analysis via multiscale gradient watershed hierarchies,” IEEE Trans. Image Processing, vol. 8, no. 1, pp. 69-79, Jan. 1999.
[16]I. E. Pratikakis, H. Sahli, and J. Cornelis, “Low level image partitioning guided by the gradient watershed hierarchy,” Signal Processing, vol. 75, no. 2, pp. 173-195, 1999.
[17]M. W. Hansen and W. E. Higgins, “Watershed-based maximum homogeneity filtering,” IEEE Trans. Image Processing, vol. 8, no. 7, pp. 982-988, July 1999.
[18]K. Saarinen, “Color image segmentation by a watershed algorithm and region adjacency graph Processing,” in Proceedings of the IEEE International Conference on Image Processing, Austin, Texas, Nov. 1994, pp. 1021-1025.
[19]F. Depiero, M. Trived and S. Serbin, “Graph matching using a direct classification of node attendance,” Pattern Recognition, vol. 29, no. 6, pp. 1031-1048, 1996.
[20]P. Colantoni and B. Laget, “Color image segmentation using region adjacency graph,” in Sixth International Conference on Image Processing and its Applications, Dublin, Ireland, July 1997, pp. 698-702.
[21]K. Haris, S. N. Efstratiadis, and N. Maglaveras, “Watershed-based image segmentation with fast region merging,” in Proceedings of the IEEE International Conference on Image Processing, Chicago, Illinois, Oct. 1998, vol. 3, pp. 338-342.
[22]Y. El-Sonbaty and M. A. Ismail, “A new algorithm for subgraph optimal isomorphism,” Pattern Recognition, vol. 31, no. 2, pp. 205-218, Jan. 1998.
[23]A. Bakowski and G. A. Jones, “Video surveillance tracking using colour region adjacency graphs,” in Seventh International Conference on Image Processing and its Applications, Manchester, UK, July 1999, vol. 2, pp. 794-798.
[24]A. Tremeau and P. Colantoni, “Regions adjacency graph applied to color image segmentation,” IEEE Trans. Image Processing, vol. 9, no. 4, pp. 735-744, Apr. 2000.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊