A Deformable Model for Complete Boundary Detection
Renato Dedić, Madjid Állili · 2006
Object recognition using the shape of objects boundaries and surface reconstruction using slice contours rely on the identification of the complete boundary information of the segmented objects in the scene. Geometric deformable models (GDM) using the level sets method provide a very efficient framework for image segmentation. However, the segmentation results provided by these models are usually dependent on the contour initialization, and in most cases where the strategy is to detect all the scene objects, the results of the segmentation will only provide partial objects boundaries. In this work, we propose a new method to detect the complete boundary information of segmented objects. This new method uses a way to keep track of already segmented parts of the image and gradient distribution analysis.