Python, image registration

Navigation
filtering	Contents

Image registration

Just like with filtering image registration handles 2D and 3D registration transparently, the only requirement is that both images to be registered are of the same dimension. Given two images test and reference, a registration optimizing a transformation t by using an optimizer opt, and using n multi-resolution levels can then be run according to

aligned_image = pymia.register_images(src=test, ref=refernce, 
                                    transform=t, optimizer=opt, mglevels=n)

Note, that the obtained transformation is currently lost. The transformation t is specified as string according to the available transformation plug-ins for 2D and 3D, and the optimizer opt can be selected from these plugins.

A secondary optimizer can be given by the refine parameter.