图像配准综述image registration.pptVIP

Introduction Medical imaging A step bring the modalities involved into spatial alignment, a procedure referred to as registration. A fusion step is required for the integrated display of the data involved. Spatial Registration Methods Extrinsic Registration Methods Characteristic: rely on artificial objects attached to the patient Advantage:objects visible accurately detectable computed explicitly no need optimization algorithms Drawback :often invasive character of the marker objects Intrinsic Registration Methods  Landmark base registration methods Technically, the identification of landmark points is a segmentation procedure, but we reserve the classification segmentation based registration for methods relating to segmentation of structures of higher order, i.e., curves, surfaces, and volumes. Intrinsic Registration Methods  Segmentation based registration methods Advantage:easy to perform the computational complexity relatively low Drawback:the limited accuracy need a good initial position Intrinsic Registration Methods  Voxel property based registration methods operate directly on the image grey values Advantage:automatic and very fast nature easy implementation Scope of application :as a coarse pre-registration in various other registration areas Intrinsic Registration Methods  full-image-content based voxel property registration methods Cross-correlation (of original images or extracted feature images) Fourier domain based cross-correlation, and phase-only correlation Minimization of variance of intensity ratios. Minimization of variance of grey values within segments Minimization of the histogram entropy of difference images Histogram clustering and minimization of histogram