In this paper, we propose a novel method for unsupervised color-texture segmentation. The approach aims at combining color and texture features and active contours to build a fully automatic segmentation algorithm. By fully automatic, we mean the steps of region initialization and calculation of the number of regions are performed automatically by the algorithm. Furthermore, the approach combines boundary and region information for accurate region boundary localization. We validate the approach by examples of synthetic and natural color-texture image segmentation. © Springer-Verlag Berlin Heidelberg 2006.
CITATION STYLE
Allili, M. S., & Ziou, D. (2006). Automatic color-texture image segmentation by using active contours. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4153 LNCS, pp. 495–504). Springer Verlag. https://doi.org/10.1007/11821045_52
Mendeley helps you to discover research relevant for your work.