Image registration metrics

D Image Registration 78 The Registration Problem 80 Similarity Metrics 81... 

In this section, we will provide a brief overview of the different models and metrics, but will leave the optimization methods for further reading. Substantive reviews of image registration can be found from several sources, and the interested reader is encouraged to read these for details. 

The main trouble with interpolation operation is that it can modify the gray level values, affecting the evaluation of the similarity function. This effect can lead to incorrect registration results if a histogram-based metric, such as Ml or NMl, is used. Often, the image dynamic is reduced to avoid this effect. 

Metrics used will also depend on the registration technique. Geometric approaches, in which expHcit models of the anatomy are used, employ point or surface matching metrics. In contrast, intensity-based approaches match intensity patterns in each image based on mathematical or statistical criteria. 

---