Difference Squared Hausdorff Distance based medical image registration

In medical image registration, Hausdorff distance has only attributed the similarity of the feature points of the images, not considered the overall information of the images, and is sensitive to disturbances. This paper adds an overall factor-the difference square between the images-in the traditional Hausdorff distance, takes the whole information of the images as a part of the metric, forms the Difference Squared Hausdorff Distance(DSHD) metric. The experiment results show that the method solves the problem the traditional Hausdorff distance has, and improves the registration precise.