Fuzzy C-means (FCM) clustering is an unsupervised clustering technique and is often used for the unsupervised segmentation of multivariate images. The segmentation of the image in meaningful regions with FCM is based on spectral information only. The geometrical relationship betweenneighbouring pi.yels is not used. In this paper, a semisupervised FCM technique is used to add geometrical information during clustering. The local neighbourhood of each pixel determines the condition of each pixel, which guides the clustering process. Segmentation experiments with the Geometrically Guided FCM (GG-FCM) show improved segmentation above traditional FCM such as more homogeneous regions and less spurious pixels.
GAIA is a preliminary concept for an astrometric mission, recently recommended within the context of ESA's 'Horizon 2000 Plus' long-term scientific programme. In its present form, the experiment is estimated to lead to positions, proper motions, and parallaxes of some 50 million objects, down to about V = 15 mag, with an accuracy better than 10 microarcsec, along with multi-colour multi-epoch photometry of each object. The scientific case for such a mission is dramatic: distances and kinematical motions for tens of millions of objects, throughout our Galaxy, would be obtained-the expected accuracy is such that direct (trigonometric) distance estimates to the galactic centre would be accurate to 10%, with transverse motions accurate to about 1 km s −1 at 20 kpc. As 'by-products', the global measurements would yield unprecedented information on the space-time metric (γ to a precision of about 1 part in 10 6 or better, close to values which might distinguish currently competing theories of gravity), angular diameters of hundreds of stars, and a vast body of information on double and multiple systems. Perhaps the most dramatic of these subsidiary goals would be the possibility of screening some 100 000 stars within 100 pc for periodic photocentric motions, which would provide the most powerful and systematic method of detecting possible planetary companions proposed to date.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.