Data Entry: Please note that the research database will be replaced by UNIverse by the end of October 2023. Please enter your data into the system https://universe-intern.unibas.ch. Thanks
Local intensity mapping for hierarchical non-rigid registration of multi-modal images using the cross-correlation coefficient
Editor(s)
Pluim, JPW; Likar, B; Gerritsen, FA
Book title (Conference Proceedings)
Biomedical image registration : third international workshop, WBIR 2006, Utrecht, The Netherlands, July 9 - 11, 2006 ; proceedings
Volume
4057
Place of Conference
Utrecht, The Netherlands
Publisher
Springer
Place of Publication
Berlin
Pages
S. 26-33
Keywords
Computer Vision -> Image Registration
Abstract
The hierarchical subdivision strategy which decomposes the non-rigid matching problem into numerous local rigid transformations is a very common approach in image registration. For multi-modal images mutual information is the usual choice for the measure of patch similarity. As already recognized in the literature, the statistical consistency of mutual information is drastically reduced when it is estimated for regions covering only a limited number of image samples. This often affects the reliability of the final registration result. In this paper we present a new intensity mapping algorithmwhich can locally transform images of different modalities into an intermediate pseudo-modality. Integrated into the hierarchical framework, this intensity mapping uses the local joint intensity histograms of the coarsely registered sub-images and allows the use of the more robust cross-correlation coefficient for the matching of smaller patches.