1 / 26

ARISTOTLE UNIVERSITY OF THESSALONIKI. DEPARTMENT OF INFORMATICS

ARISTOTLE UNIVERSITY OF THESSALONIKI. DEPARTMENT OF INFORMATICS 2D/3D Image Registration and Alignment: A Review Stelios Krinidis Presentation outline Definitions General aspects ICP algorithm Shape-based algorithm References Definitions Registration: a fundamental task in image

Télécharger la présentation

ARISTOTLE UNIVERSITY OF THESSALONIKI. DEPARTMENT OF INFORMATICS

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. ARISTOTLE UNIVERSITY OF THESSALONIKI.DEPARTMENT OF INFORMATICS 2D/3D Image Registration and Alignment: A Review Stelios Krinidis Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  2. Presentation outline • Definitions • General aspects • ICP algorithm • Shape-based algorithm • References Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  3. Definitions • Registration:a fundamental task in image • processing used to match two or more pictures taken, • for example, at different times, from different sensors, • or from different viewpoints. • Alignment:a fundamental task in image processing • used to match two or more pictures that are similar • but not alike, for example different sections from a 3D • object. Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  4. General aspects (1) • Registration/Alignment can be used to: • 3D object reconstruction from its 2D sections. • 3D object visualization and morphological analysis. • Compare medical tissues (taken at different times) • showing tumor growth, internal abnormalities, etc. • Medical and surgical analysis, tests and simulations. Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  5. General aspects (2) • Registration/Alignment (2D and 3D) compensation: • rotation and translation (MRI, CT, etc) • non-rigid transforms (physical sectioning of • biological tissues, anatomical atlases, etc) Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  6. General aspects (3) • Proposed Registration/Alignment methods: • fiducial marker-based • feature-based using contours • crest lines or characteristics points • gray level-based Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  7. Iterative Closest Point (ICP) • It can be used with the following representations of • geometrical data: • points sets • line segments (polylines) • implicit curves: g(x,y,z) = 0 • parametric curves: (x(u),y(u),z(u)) • triangle sets (faceted surfaces) • implicit surfaces: g(x,y,z) = 0 • parametric surfaces: (x(u,υ),y(u,υ),z(u,υ)) Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  8. Iterative Closest Point (ICP) • Characteristics: • monotonic convergence to the nearest local minimum • rapid convergence during the first few iterations • global convergence depends on the initial parameters Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  9. Iterative Closest Point (ICP) Model point set: Data point set: Closest point set: Distance metric: Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  10. Iterative Closest Point (ICP) Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  11. Iterative Closest Point (ICP) Quaternion is the eigenvector related to the largest eigenvalue: Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  12. Iterative Closest Point (ICP) Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  13. Iterative Closest Point (ICP) Point Set Matching Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  14. Iterative Closest Point (ICP) Curve Set Matching Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  15. Iterative Closest Point (ICP) Surface Set Matching Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  16. Shape-Based Alignment • Alignment of 2D serially acquired sections forming a • 3D object • Characteristics: • shape-based algorithm (contours) • global energy function (expressing similarity between • neighboring slices). • no direction is privileged • no global offset • no error propagation Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  17. Shape-Based Alignment N : frame number Nx: horizontal image dimension Ny: vertical image dimension R : neighborhood’s length f : pixel similarity metric Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  18. Shape-Based Alignment Di : Distance Transform of image i Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  19. Shape-Based Alignment Distance Transform: each pixel has value equal to the pixel’s distance from the nearest non-zero pixel. Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  20. Shape-Based Alignment Alignment Errors Statistics Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  21. Shape-Based Alignment Alignment Errors Statistics Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  22. Shape-Based Alignment Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  23. Shape-Based Alignment Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  24. Shape-Based Alignment Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  25. References • P. Van den Elsen, E.J.D. Paul, and M.A.Viergever. Medical Image Matching – A review with classification. IEEE engineering in Medicine and Biology, 12(1):26-39, 1993. • M.J.Besl and N.McKay. A Method for the Registration of 3D Shapes. IEEE transactions of Pattern Analysis and Machine Intelligence(PAMI), 14(2):239-256, 1992 • G.Borgefors. Hierarchical Chamfer Matching: A parametric edge matching algorithm. IEEE transactions of Pattern Analysis and Machine Intelligence(PAMI), 679-698, 1986. • W.Wells III, P.Viola, H.Atsumi, S.Nakajima, and R.Kikinis. Multimodal volume registration by maximization of mutual information. Medical Image Analysis, 1(1):33-51, 1996. Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

  26. References • C.Nikou, J.P.Armspach, F.Heitz, I.J.Namer, and D.Grucker. MR/MR and MR/SPECT registration of brain by fast stochastic optimization of robust voxel similarity measures NeuroImage, 8(1):30-43, 1998. • S.Krinidis, N.Nikolaidis, I.Pitas. Shape Based Alignment of 3-D Volume Slices. International Conference on Electronics, Circuits and Systems (ICECS'00) Kaslik, Lebanon, 17-20 September 2000. Department of Informatics, Aristotle University of Thessaloniki, May 4, 2001

More Related