scalismo

Contains classes for representing discrete and continuous images as well as filters for filtering both types of images.

In this library, the standard type of images are just functions, defined on some domain. There is also the more traditional discrete image type, which represents image data on a regular pixel/voxel grid.

DiscreteImages are mainly used for reading/writing and simple manipulation of image values. The continuous representation of images is much more flexible and most method for manipulating images are defined only on this image type. A discrete image can be converted to a continuous image by using an interpolation procedure:

val domain = StructuredPoints(Point(0,0), Vector(1,1), Index(255,255))
val di = DiscreteImage(domain)(0)
val discreteImage =  DiscreteImage(domain, (_ : Point[_2D]) => 1.0f)
val continuousImage = discreteImage.interpolate(3)

To get back the discrete representation, we can sample the image values on a regular grid:

val newDomain = StructuredPoints(Point(0,0), Vector(1,1), Index(128,128))
val resampledDiscreteImage = continuousImage.sample(domain, 0)

Contains classes for handling datasets of registered meshes and building PCA model out of them.

The most important class of this package is the scalismo.statisticalmodel.dataset.DataCollection.