A new minimum variance region growing algorithm for image segmentation

• DOI: 10.1016/s0167-8655(97)00012-3
• Publication Date: 2003-05-12T22:21:32Z
• Authors: Chantal Revol; Michel Jourlin
• Keywords: 0301 basic medicine; Image segmentation Region growing Homogeneity criterion 3D MR imaging Pattern Recognition Letters 23 (2002) 137±150 www.elsevier.com/locate/patrec; Image segmentation; [SPI]Engineering Sciences [physics]; 03 medical and health sciences; Region growing; 616; [INFO]Computer Science [cs]; Homogeneity criterion; 3D MR imaging Pattern Recognition Letters 23 (2002) 137±150 www.elsevier.com/locate/patrec; 004

Abstract Region growing is a very useful technique for image segmentation. Its efficiency mainly depends on its aggregation criterion. In the present paper, a new algorithm is proposed with a homogeneity criterion based on an adequate tuning between spatial neighbourhood and histogram neighbourhood. It differs from other techniques by reconsidering the pixel (or voxel) assignments on each step by a process which minimizes variance through special dilations. Thus, the region created by an initial seed can be non-connected and possibly does not contain this seed. Examples are given in dental surgery for 2D X-Ray images (and their associated 3D block) and for 3D images acquired by the Morphometre, the new 3D scanner constructed by GEMSE (General Electric Medical Systems).