This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1 Non-rigid multi-modal image registration using cross-cumulative residual entropy Fei Wang 1 Baba C. Vemuri 2 1 IBM Almaden Research Center, 2 Department of CISE 650 Harry Road, San Jose, CA 95120 University of Florida, Gainesville, FL 32611 In this paper we present a new approach for the non-rigid registration of multi-modality images. Our approach is based on an information theoretic measure called the cumulative residual entropy (CRE), which is a measure of entropy de ned using cumulative distributions. Cross-CRE between two images to be registered is de ned and maximized over the space of smooth and unknown non-rigid transformations. For ef cient and robust computation of the non-rigid deformations, a tri-cubic B-spline based representation of the deformation function is used. The key strengths of combining CCRE with the tri-cubic B-spline representation in addressing the non-rigid registration problem are that, not only do we achieve the robustness due to the nature of the CCRE measure, we also achieve computational ef ciency in estimating the non-rigid registration. The salient features of our algorithm are: (i) it accommodates images to be registered of varying contrast+brightness, (ii) faster convergence speed compared to other information theory-based measures used for non-rigid registration in literature, (iii) analytic computation of the gradient of CCRE with respect to the non-rigid registration parameters to achieve ef cient and accurate registration, (iv) it is well suited for situations where the source and the target images have eld of views with large non-overlapping regions. We demonstrate these strengths via experiments on synthesized and real image data. Key words: Information theory, Shannon Entropy, Multi-modal non-rigid registration, B-splines. I. I NTRODUCTION Image registration is a ubiquitous problem in medical imaging and many other applications of image analysis including but not limited to geo-spatial imaging, satellite imaging, movie editing, archeology etc. In medical imaging, non-rigid registration is particularly common in longitudinal studies such as in child development, ageing studies and also in comparisons between controls and pathologies to assess progress or remission of disease. There is an abundance of non-rigid registration algorithms in literature, the most popular approaches come in two varieties, those that assume brightness constancy in their cost function being optimized and others that use information theory based cost functions that don't require the aforementioned restrictive assumption. The former are applicable only to same modality data sets while the latter can be applied to multi-modal data sets. There are many applications wherein use of multi-modality data sets is desired e.g., image-guided neurosurgery where an MR is used to locate 2 the tumor and a registered high resolution CT is used for guidance. Another application is in cognitive studies where, MRI and fMRI registration is sought....
View Full Document
This note was uploaded on 01/22/2012 for the course COP 5615 taught by Professor Staff during the Fall '08 term at University of Florida.
- Fall '08