7 February 1997 Method for multispectral image segmentation based on the shape of the color clusters
Author Affiliations +
Proceedings Volume 2949, Imaging Sciences and Display Technologies; (1997) https://doi.org/10.1117/12.266349
Event: Advanced Imaging and Network Technologies, 1996, Berlin, Germany
The paper describes an algorithm for multi-spectral images segmentation that takes into account the shape of the clusters formed by the pixels of the same object in the spectral space. The expected shape of the clusters is based on the dichromatic reflection model, and its extension for optically homogeneous materials. Further the influence of the illumination and image formation by a color CCD camera are considered. Based on expected shape of clusters we propose a criterion of similarity/homogeneity for the extended region merging algorithm. This criterion works successfully in case of objects of voluntary shape and illumination by one or several sources of the same spectrum.
© (1997) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Natalia H. Kroupnova, Natalia H. Kroupnova, "Method for multispectral image segmentation based on the shape of the color clusters", Proc. SPIE 2949, Imaging Sciences and Display Technologies, (7 February 1997); doi: 10.1117/12.266349; https://doi.org/10.1117/12.266349

Back to Top