Registration is a decisive, primary step in image analysis that helps to obtain absolute information by combining multiple data sources. This pre-processing task is one of the most essential measures in medical images making them useful for different applications such as classification, change detection and image fusion. With the advent of multiple modalities that yield numerous images, registering them becomes a challenging issue. Conventional approaches for image registration incident a meagre performance due to their vulnerability in scale and intensity variations. In this paper we propose an optimized FUZZY- SIFT Matching technique for image registration. Initially Scale Invariant Feature Transform (SIFT) is applied to extract key points from images. Images are segmented to regions based on Fuzzy C-means clustering approach which produces clusters. Key points are matched based on their gradient orientations from the clusters of both reference image and target image and finally image warping is performed by applying piecewise linear transformation function. Experimental results indicate that the proposed method improves the match performance compared to other usual methods in terms of correct-match rate and aligning accuracy.

1. Barbara et al, (2003) "Image Registration methods: a survey", Image and Vision Computing, Vol. 21, No. 11, pp. 977-1000.
2. Collignon, A. , Vandermuelen, D. , Suetens, P. and Marchal, G. (1995b) 3d multi-modality medical image registration using feature space clustering. In Ayache, N. (ed. ), Computer Vision, Virtual Reality and Robotics in Medicine, pp. 195–204. Springer Verlag, Berlin.
3. Damas, S. , Cordo?n, O. , Santamari?a, J. , 2011. Medical Image Registration using Evolutionary, Computational Intelligence Magazine, IEEE, Vol. 6, 26-42. J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed. , vol. 2. Oxford: Clarendon, 1892, pp. 68–73.
4. Joaquim Salvi , Carles Matabosch, David Fofi , Josep Forest. A review of recent range image registration methods with accuracy evaluation Image and Vision Computing. May 16 (2006).
5. S. K. Kyriacou, C. Davatzikos, J. S. Zinreich, and N. R. Bryan, "Non –linear elastic registration of brain images with tumor pathology using a biomechanical model", IEEE Trans. Med. Imag. , Vol. 18, no. 7, pp. 580-592. , Jul. 1999.
6. Lemieux, L. , Wieshmann, U. , Moran, N. , Fish, D. , 1998. The detection and significance of subtle changes in mixed-signal brain lesions by serial MRI scan matching and spatial normalization, Med. Image Anal. 2, 227–242.
7. Linsker, R. (1986) From basic network principles to neural architecture. Proc. Natl Acad. Sci. USA, 83, 7508–7512, 8390– 8394, 8779–8783.
8. Maes, F. , Vandermeulen, D. , Suetens, P. , 2003, Medical image registration using mutual information, IEEE, Vol. 91, 1699-1722.
9. Martin, S. , Durrani, T. S. , 2007. A New Divergence Measure for Medical Image Registration, IEEE Transaction on Image Processing, Vol. 16, 957 – 966.
10. Pelizzari, C. , Chen, G. , Spelbring, D. , Weichselbaum, R. and Chen, C. (1989) Accurate three dimensional registration of CT, PET and or MR Images of the brain. J. Comp. Assis. Tomogr. , 13, 20-26.
11. Pluim, J. P. W. , Maintz, J. B. A, Viergever, M. A. , 2004. F-information measures in medical registration, IEEE Transactions on Medical Imaging, Vol. 23, 1508-1516.
12. T. Rohlfing, c. R. Maurer, Jr. , D. A>Bluemke, and M. A. Jacobs, "Volume-preserving non rigid registration of MR breast images using free-form deformation withan incompressibility constraint", IEEE Trans. Med. Imag. , Vol. 22, no. 6, pp. 730-741. , Jun. 2003.
13. M. Staring, U. A. Van der Heide, S. Klein, M. A. Viergever, and J. P. W. Plumin, "Registration of cervical MRI using multifeature mutual information", IEEE Trans. Med. Imag. , Vol. 28, no. 9, pp. 1412-1421. , sep. 2009.
14. Viola, P. andWells,W. (1995) Alignment by maximization of mutual information. In Proc. 5th Int. Conf. Computer Vision, Boston, IEEE.
15. Wachowiak, M. P. , Peters, T. M. , 2006. High performance medical registration using new optimization techniques, IEEE Transactions on Information Technology in Biomedicine, Vol. 10, pp. 344-353.

Medical Images Registration Sift Clustering Dog Filter Piecewise Linear Transformation