A method of sift feature points matching for image mosaic

This paper presents a approach of SIFT feature points matching for image mosaic. This method combines improved K-means clustering and simulated annealing algorithm to match SIFT feature points. Firstly, high robust points are extracted by SIFT algorithm; Secondly, cluster with the initial centers obtained by density function, and then optimize the results of clustering which are used as initial results of simulated annealing algorithm by perturbation; Thirdly, match feature points according to Nearest Neighbor algorithm; Finally, calculate the homography and realize image mosaic. This method does not need to traverse all feature points and avoid trapping in a local extremum. Experimental results prove that the method is only relative to geometric position of feature points, and is robust on scale invariant, arbitrary rotation and scaling.

[1]  Anil K. Jain,et al.  Pores and Ridges: High-Resolution Fingerprint Matching Using Level 3 Features , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

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

[3]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[4]  R. J. Althof,et al.  A rapid and automatic image registration algorithm with subpixel accuracy , 1997, IEEE Transactions on Medical Imaging.

[5]  Yiu-ming Cheung,et al.  k*-Means: A new generalized k-means clustering algorithm , 2003, Pattern Recognit. Lett..

[6]  Sujoy Ghose,et al.  Point matching using asymmetric neural networks , 1993, Pattern Recognit..

[7]  Nilanjan Ray,et al.  Pattern Recognition Letters , 1995 .

[8]  Sunil Arya,et al.  An optimal algorithm for approximate nearest neighbor searching fixed dimensions , 1998, JACM.

[9]  Claire Cardie,et al.  Proceedings of the Eighteenth International Conference on Machine Learning, 2001, p. 577–584. Constrained K-means Clustering with Background Knowledge , 2022 .

[10]  Robert C. Bolles,et al.  Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[11]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[12]  Yi Chen,et al.  Pores and Ridges: High-Resolution Fingerprint Matching Using Level 3 Features , 2007 .

[13]  B. V. K. Vijaya Kumar,et al.  A Bayesian Approach to Deformed Pattern Matching of Iris Images , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Zhengwei Yang,et al.  Image registration and object recognition using affine invariants and convex hulls , 1999, IEEE Trans. Image Process..

[15]  Nasser M. Nasrabadi,et al.  Hopfield network for stereo vision correspondence , 1992, IEEE Trans. Neural Networks.

[16]  Matthijs C. Dorst.  Distinctive Image Features from Scale-Invariant Keypoints , 2011 .

[17]  Sunil Arya,et al.  Approximate nearest neighbor queries in fixed dimensions , 1993, SODA '93.

[18]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[19]  B. S. Manjunath,et al.  Registration Techniques for Multisensor Remotely Sensed Imagery , 1996 .

[20]  Richard A. Baldock,et al.  Robust Point Correspondence Applied to Two-and Three-Dimensional Image Registration , 2001, IEEE Trans. Pattern Anal. Mach. Intell..