Title :
Active contour driven by local region force
Author_Institution :
Sch. of Inf. Sci. & Eng., Shandong Univ., Jinan, China
Abstract :
Active contour models is one of the most popular and successful segmentation methods. In this paper we propose a new geodesic active contour model combination with an external force field which is based on region information inside and outside the curves. The external region force based on region information, which can be divided into global information and local information, is incorporated into the geodesic active contour model to drive the curves to Converge to the boundaries of the objects. Our proposed method then is incorporated into the level set foundational to compute the gradient flow. Due to the combination of the geodesic model and region information, our proposed method can get better performance both in synthetic images and real images.
Keywords :
computational geometry; differential geometry; image segmentation; external force field; geodesic active contour model; gradient flow; local region force; segmentation methods; Artificial intelligence; Capacitance-voltage characteristics; Force; Image recognition; Image segmentation; Geodesic active contour model; Region force; level set functional; region information;
Conference_Titel :
Communication Software and Networks (ICCSN), 2011 IEEE 3rd International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-1-61284-485-5
DOI :
10.1109/ICCSN.2011.6014676