Abstract: An unsupervised method of segmenting data sets using a region growing technique in which data points are initially assigned to a single class, new classes are seeded and points in the data set tested by calculating the probability that they belong to the new class. The probability distributions used in the calculation are adapted as points are reassigned. Classes which fail to grow are discarded. The technique may be applied to the segmentation of data sets in which the data points are taken from medical images. The method may be applied to the demarcation of different parts of structures, e.g. in the medical field demarcating an aneurysm from the surrounding blood vessels in an image or 3-D model of a patient's vasculature. The method may involve using a shape descriptor which is representative of the shape of the structure at each point under consideration. Thus the different parts are distinguished on the basis of their shape.

Abstract: A method of calibrating a linear pushbroom imaging device used in non-destructive testing of a part manufactured according to a CAD model involves identifying the location of fixed reference points, such as tooling balls in the CAD model. Positioning the part on a fixture having tooling balls fixed relative to each other in the same manner as those in the CAD model. Adjusting the part on the fixture such that points on the surface of the part known to be accurate have the same locations relative to the tooling bails as the CAD part is to the CAD tooling balls. Obtaining several projection images of the part and tooling balls at different angles about an axis of rotation passing through the part. Using the known 3D locations of the center of the CAD tooling balls and the corresponding measured screen locations of the is actual tooling balls, transformation matrices, G, M.sub.

Abstract: An apparatus and method of classifying and sorting by shape crystalline objects such as synthetic diamonds in which an image of the object taken from an angle defined in relation to the object is compared to one or more templates in order to characterize the object.