在影像辨識的領域及數位版權管理中的浮水印，局部不變性特徵點已是被廣泛使用的方法，而我們採用尺寸不變性特徵轉換法來截取不變性特徵點，此方法所取出的特徵點用於不同影像之間的辨識具有相當的可靠性，並且，這些特徵點能抵抗影像轉檔、尺寸縮放、旋轉等變化；然而，大量的特徵點對辨識階段將是一個重要的問題，若資料庫中的影像增加時，數量的問題會將會更加惡化，因此，我們提出一個有效率的索引方法，以搜尋樹為基礎，以降低在辨識時所需的特徵點數量及降低總共的詢問時間。

Local invariant features method is one of the popular methods in image recognition and in Digital Right Management (DRM). In this work, we propose to use the Scale Invariant Feature Transform (SIFT) to generate local invariant feature keypoints, which are reliable in matching among different images, for DRM. The feature keypoints extracted by the SIFT are invariant to image translation, scaling, rotation, etc., and hence a promising technique to identify plagiarism among images. However, the large number of local invariant feature keypoints extract by the SIFT poses a major problem in the matching stage. This situation deteriorates as the number of images to be matched in the database increases. In this work, we present efficient indexing methods based on search trees to reduce the number of keypoint matching required and to reduce the total query time.

[1] Demo Software: SIFT Keypoint detector, http://www.cs.ubc.ca/~lowe/keypoints.
[2] J. S. Beis and D. G. Lowe, “Shape Indexing Using Approximate
Nearest-Neighbour Search in High-Dimensional Spaces”, Proceedings of the
1997 Conference on Computer Vision and Pattern Recognition, p.1000, June
17-19, 1997.
[3] J. H. Freidman , J. L. Bentley , and R.A. Finkel, “An Algorithm for Finding Best Matches in Logarithmic Expected Time”, ACM Transactions on Mathematical Software, v.3 n.3, p.209-226, September, 1977.
[4] K. Grauman and T. Darrell, “Efficient Image Matching with Distributions of Local Invariant Features”, Proceedings of the 2005 IEEE Computer Society
Conference on Computer Vision and Pattern Recognition, v.2, p.627-634, June
20-26, 2005.
[5] W. Grimson and D. Huttenlocher, “On the sensitivity of geometric hashing”. In Proceedings 3rd International Conference on Computer Vision, p.334–338, 1990.
[6] Y. Lamdan, J. Schwartz, and H. Wolfson, “Affine invariant model based object recognition”. IEEE Transactions on Robot Automn, v.6 n.5, p.578–589,1990.
[7] Y. Lamdan and H. Wolfson, “Geometric hashing: a general and efficient
model-based recognition scheme”. In Proceedings 2nd International Conference on Computer Vision, p.238–249, 1988.
[8] H. Y. Lee, H. Kim, and H.K. Lee, “Robust image watermarking using local
invariant Features”, Optical Engineering, v.45 n.3, 037002, March, 2006.
[9] D. G. Lowe, “Distinctive Image Features from Scale-Invariant Keypoints”, International Journal of Computer Vision, v.60 n.2, p.91-110, November, 2004.
[10] D. G. Lowe, “Object Recognition from Local Scale-Invariant Features”,
Proceedings of the International Conference on Computer Vision, v.2, p.1150,
September 20-25, 1999.
[11] L. Miclet and M. Dabouz, “Approximative fast nearest neighbor recognition”, Pattern Recognition Letters, v.1, p.277–285, 1983.
[12] R. Sproull. “Refinements to nearest-neighbor searching in k-dimensional trees”, Algorithmica, v.6, p.579–589, 1991.