Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

A Bayesian Approach for Image Segmentation with Shape Priors

Conference ·
OSTI ID:957402

Color and texture have been widely used in image segmentation; however, their performance is often hindered by scene ambiguities, overlapping objects, or missingparts. In this paper, we propose an interactive image segmentation approach with shape prior models within a Bayesian framework. Interactive features, through mouse strokes, reduce ambiguities, and the incorporation of shape priors enhances quality of the segmentation where color and/or texture are not solely adequate. The novelties of our approach are in (i) formulating the segmentation problem in a well-de?ned Bayesian framework with multiple shape priors, (ii) ef?ciently estimating parameters of the Bayesian model, and (iii) multi-object segmentation through user-speci?ed priors. We demonstrate the effectiveness of our method on a set of natural and synthetic images.

Research Organization:
Ernest Orlando Lawrence Berkeley National Laboratory, Berkeley, CA (US)
Sponsoring Organization:
Life Sciences Division
DOE Contract Number:
AC02-05CH11231
OSTI ID:
957402
Report Number(s):
LBNL-1891E
Country of Publication:
United States
Language:
English

Similar Records

Variational and Shape Prior-based Level Set Model for Image Segmentation
Journal Article · Thu Sep 30 00:00:00 EDT 2010 · AIP Conference Proceedings · OSTI ID:21428592

Flexible prior models in Bayesian image analysis
Conference · Tue Dec 31 23:00:00 EST 1991 · OSTI ID:7133957

Flexible prior models in Bayesian image analysis
Conference · Mon Nov 30 23:00:00 EST 1992 · OSTI ID:10112156