Noblet Vincent, Heinrich Christian, Heitz Fabrice, Armspach Jean-Paul
Laboratoire des Sciences de l'Image, de l'Informatique et de la Télédétection, LSIIT, UMR CNRS-ULP 7005, Bd Sébastien Brant, 67412 Illkirch, France.
Med Image Comput Comput Assist Interv. 2008;11(Pt 2):897-904. doi: 10.1007/978-3-540-85990-1_108.
Image registration aims at estimating a consistent mapping between two images. Common techniques consist in choosing arbitrarily one image as a reference image and the other one as a floating image, thus leading to the estimation of inconsistent mappings. We present a symmetric formulation of the registration problem that maps the two images in a common coordinate system halfway between them. This framework has been considered to devise an efficient strategy for mapping a large set of images in a common coordinate system. Some results are presented in the context of 3-D nonrigid brain MR image registration for the construction of average brain templates.
图像配准旨在估计两幅图像之间的一致映射。常见技术是任意选择一幅图像作为参考图像，另一幅作为浮动图像，从而导致估计出不一致的映射。我们提出了一种配准问题的对称公式，该公式在两幅图像之间的公共坐标系中将它们进行映射。这个框架已被用于设计一种在公共坐标系中映射大量图像的有效策略。在用于构建平均脑模板的三维非刚性脑磁共振图像配准的背景下给出了一些结果。
Zotero 插件