期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2013
卷号：4
期号：5
页码：692-695
出版社：TechScience Publications
摘要：In this paper we present a method for unsupervised image clustering and B-spline curve interpolation for remove the noise. Images are clustered such that the mutual information between the clusters and the image content is maximally preserved. The clustering principle is applied to both discrete and continuous image representations, using Gaussian densities. We find that whole-brain analysis in this manner allows automatic classification of images based on Disease if the whole brain is included, If a sequence of planar points and vectors are given then a free-form curve can calculated which interpolates the points and has the given tangent vectors in these points. Our method gives a fast interpolation of these data using extra control points. Then we provide a method which allows to interpolate the same set of data without any predefined order of the points, i.e. a set of scattered points with the vectors.
关键词：Unsupervised image clustering; B-spline curve;interpolation; mixture of Gaussians
