New efficient image segmentation algorithm: competitive region growing of initial regions

Presents a new simple and efficient region growing algorithm and its superior results against the current medical region growing practice. The algorithm uses initial regions and expands them to their natural borders. If all the required regions can be initially enumerated and roughly located the algorithm does not need any control parameters and gives optimal result. The algorithm was designed for both interactive and knowledge-based segmentation of volumetric medical images.

[1]  Y. Louhisalmi,et al.  Image transformation from polar to Cartesian coordinates simplifies the segmentation of brain images , 1992, 1992 14th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[2]  Eugene C. Freuder Affinity: A relative approach to region finding , 1976 .

[3]  Steven W. Zucker,et al.  Region growing: Childhood and adolescence* , 1976 .