Controlled Line Smoothing by Snakes

A major focus of research in recent years has been the development of algorithms for automated line smoothing. However, combination of the algorithms with other generalization operators is a challenging problem. In this research a key aim was to extend a snakes optimization approach, allowing displacement of lines, to also be used for line smoothing. Furthermore, automated selection of control parameters is important for fully automated solutions. An existing approach based on line segmentation was used to control the selection of smoothing parameters dependent on object characteristics. Additionally a new typification routine is presented, which uses the same preprocessed analysis for the segmentation of lines to find suitable candidates from curve bends. The typification is realized by deleting undersized bends and emphasizing the remaining curve bends. The main results of this research are two new algorithms for line generalization, where the importance of the line smoothing algorithm lies in the usage of a optimization approach which can also be used for line displacement.

[1]  Robert B McMaster,et al.  The Integration Of Simplification And Smoothing Algorithms In Line Generalization , 1989 .

[2]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[3]  Robert B Mc Master,et al.  Generalization in Digital Cartography Resource Publications in Geography , 1992 .

[4]  Lars Harrie,et al.  An Optimisation Approach to Cartographic Generalisation , 2001 .

[5]  Emmanuel Fritsch,et al.  Spectral Representations of Linear Features for Generalisation , 1995, COSIT.

[6]  Stefan Steiniger,et al.  Snakes: a technique for line smoothing and displacement in map generalisation , 2004 .

[7]  Monika Sester.  GENERALIZATION BASED ON LEAST SQUARES ADJUSTMENT , 2000 .

[8]  J. L. G. Balboa,et al.  Frequency Filtering of Linear Features by Means of Wavelets. A Method and an Example , 2000 .

[9]  Byron Nakos,et al.  Local length ratio as a measure of critical points detection for line simplification , 2002 .

[10]  Eric Saux,et al.  B-spline Functions and Wavelets for Cartographic Line Generalization , 2003 .

[11]  Andrew P. Witkin,et al.  Uniqueness of the Gaussian Kernel for Scale-Space Filtering , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Joachim Bobrich.  Ein neuer Ansatz zur kartographischen Verdrängung auf der Grundlage eines mechanischen Federmodells , 1996 .

[13]  Keith C. Clarke,et al.  EMPIRICAL COMPARISON OF TWO LINE ENHANCEMENT METHODS , .

[14]  P. Højholt.  Solving Space Conflicts in Map Generalization: Using a Finite Element Method , 2000 .

[15]  Tapani Sarjakoski,et al.  Generalisation of vector data sets by simultaneous least squares adjustment , 2000 .

[16]  Frank Schwarzbach.  Untersuchungen zur rechnergestützten Linienglättung , 1994 .

[17]  Anne Ruas,et al.  Experiments with Learning Techniques for Spatial Model Enrichment and Line Generalization , 1998, GeoInformatica.

[18]  C. Plazanet.  Enrichissement des bases de données géographiques : analyse de la géométrie des objets linéaires pour la généralisation cartographique (application aux routes) , 1996 .