Euclidean distance between two images:

Euclidean distance between 3D images:

By default, images are center aligned:

Align the top-left corner of the first image with the center of the second image:

Align the center top of the first image and right center of the second image:

Specify positions using the standard coordinate system:

Specify positions using scaled coordinates:

Specify the alignment position for 3D images:

Construct a similarity graph of a set of images based on the earth mover distance:

Compute a matrix of distances for all image pairs:

Threshold the distances to construct an adjacency matrix and visualize using a graph:

Register two images related by a translation and a rotation:

The distance between an image and itself is always zero:

The distance between two different images may be 0:

The distance between two images is symmetric:

The distance may not be symmetric with masks:

When the images have no overlap, the distance is indeterminate:

The distance between an image and its rotations:

Normalized mutual information variation of an MRI and a translated PET image:

Correcting for overall change in brightness is a typical preprocessing with some distances:

Without correction, it is less apparent that the two images are aligned:

Some distance measures are effectively invariant to specific transformations.

Spatial reflection:

Color negation:

Pointwise addition:

Pointwise multiplication by a positive number:

Pointwise affine transformation with positive slope coefficient:

For a rectangular mask, image distance is equal to the distance of the trimmed images: