An improved SIFT matching algorithm based on locality preserving projection LPP

The problems of high dimensional feature vectors lead to large data calculating and time consuming in Scale Invariant Feature Transform (SIFT) feature matching. This paper combines the SIFT and Locality Preserving Projection (LPP) as the LPP-SIFT algorithm. First, use the SIFT algorithm to extract feature point vectors, then by the linear LPP dimensionality reduction method, reduce the amount of data. Use the Euclidean distance similarity criteria to achieve the coarse matching feature point vectors. Then re-match each pair of the match points by the way of the Gaussian neighborhood weighted average to get rid of the mismatch points. The experimental results show that the improved LPP-SIFT algorithm can reduce the amount of data, enhance the real-time and has a high matching accuracy and matching efficiency.

[1]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[2]  Cordelia Schmid,et al.  A Performance Evaluation of Local Descriptors , 2005, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  David G. Lowe,et al.  Object recognition from local scale-invariant features , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[4]  Tony Lindeberg,et al.  Feature Detection with Automatic Scale Selection , 1998, International Journal of Computer Vision.

[5]  Xiaofei He,et al.  Locality Preserving Projections , 2003, NIPS.

[6]  Hans P. Morevec.  Towards automatic visual obstacle avoidance , 1977, IJCAI 1977.

[7]  Hans P. Moravec.  Towards Automatic Visual Obstacle Avoidance , 1977, IJCAI.

[8]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[9]  Mikhail Belkin,et al.  Laplacian Eigenmaps for Dimensionality Reduction and Data Representation , 2003, Neural Computation.

[10]  Christopher G. Harris,et al.  A Combined Corner and Edge Detector , 1988, Alvey Vision Conference.