A Fast Optimization Method for Level Set Segmentation
2009 (English)In: Image Analysis: 16th Scandinavian Conference, SCIA 2009, Oslo, Norway, June 15-18, 2009. Proceedings / [ed] A.-B. Salberg, J.Y. Hardeberg, and R. Jenssen, Springer Berlin/Heidelberg, 2009, 400-409 p.Conference paper (Refereed)
Level set methods are a popular way to solve the image segmentation problem in computer image analysis. A contour is implicitly represented by the zero level of a signed distance function, and evolved according to a motion equation in order to minimize a cost function. This function defines the objective of the segmentation problem and also includes regularization constraints. Gradient descent search is the de facto method used to solve this optimization problem. Basic gradient descent methods, however, are sensitive for local optima and often display slow convergence. Traditionally, the cost functions have been modified to avoid these problems. In this work, we instead propose using a modified gradient descent search based on resilient propagation (Rprop), a method commonly used in the machine learning community. Our results show faster convergence and less sensitivity to local optima, compared to traditional gradient descent.
Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2009. 400-409 p.
Lecture Notes in Computer Science, ISSN 0302-9743 (print), 1611-3349 (online) ; 5575
Image segmentation - level set method - optimization - gradient descent - Rprop - variational problems - active contours
Engineering and Technology
IdentifiersURN: urn:nbn:se:liu:diva-19313DOI: 10.1007/978-3-642-02230-2_41ISI: 000268661000041ISBN: 978-3-642-02229-6 (print)ISBN: 978-3-642-02230-2 (online)OAI: oai:DiVA.org:liu-19313DiVA: diva2:224261
16th Scandinavian Conference on Image Analysis, June 15-18 2009, Oslo, Norway