Automatically Building 2D Statistical Shapes Using the Topology Preservation Model GNG

Image segmentation is very important in computer based image interpretation and it involves the labeling of the image so that the labels correspond to real world objects. In this study, we utilise a novel approach to automatically segment out the ventricular system from a series of MR brain images and to recover the shape of hand outlines from a series of 2D training images. Automated landmark extraction is accomplished through the use of the self-organising model the growing neural gas (GNG) network which is able to learn and preserve the topological relations of a given set of input patterns without requiring a priori knowledge of the structure of the input space. The GNG based method is compared to other self-organising networks such as Kohonen and Neural Gas (NG) maps and results are given showing that the proposed method preserves accurate models.

[1]  Nasser M. Nasrabadi,et al.  Vector quantization of images based upon the Kohonen self-organizing feature maps , 1988, ICNN.

[2]  Kalle Åström,et al.  Minimizing the description length using steepest descent , 2003, BMVC.

[3]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[4]  Hildur Ólafsdóttir,et al.  Adding Curvature to Minimum Description Length Shape Models , 2003, BMVC.

[5]  Timothy F. Cootes,et al.  A minimum description length approach to statistical shape modeling , 2002, IEEE Transactions on Medical Imaging.

[6]  José García Rodríguez,et al.  Automatic Landmarking of 2D Medical Shapes Using the Growing Neural Gas Network , 2005, CVBIA.

[7]  Timothy F. Cootes,et al.  Training Models of Shape from Sets of Examples , 1992, BMVC.

[8]  T. Martínez,et al.  Competitive Hebbian Learning Rule Forms Perfectly Topology Preserving Maps , 1993 .

[9]  Hamid Soltanian-Zadeh,et al.  Automatic landmark extraction from image data using modified growing neural gas network , 2003, IEEE Transactions on Information Technology in Biomedicine.

[10]  Helge Ritter,et al.  Topology conserving mappings for learning motor tasks , 1987 .

[11]  Thomas Martinetz,et al.  Topology representing networks , 1994, Neural Networks.

[12]  Helge J. Ritter,et al.  Three-dimensional neural net for learning visuomotor coordination of a robot arm , 1990, IEEE Trans. Neural Networks.

[13]  T. Sejnowski,et al.  A unifying measure for neighbourhood preservation in topographic mappings , 1997 .

[14]  Edwin R. Hancock,et al.  Learning mixtures of point distribution models with the EM algorithm , 2003, Pattern Recognit..

[15]  David C. Hogg,et al.  Learning Flexible Models from Image Sequences , 1994, ECCV.

[16]  Bernd Fritzke,et al.  A Growing Neural Gas Network Learns Topologies , 1994, NIPS.

[17]  Teuvo Kohonen,et al.  In: Self-organising Maps , 1995 .