Multi-Object Geodesic Active Contours (MOGAC): A Parallel Sparse-Field Algorithm for Image Segmentation

Show simple item record

dc.contributor.author Taylor, Russell
dc.contributor.author Kazhdan, Michael
dc.contributor.author Lucas, Blake
dc.date.accessioned 2012-02-28T01:02:21Z
dc.date.available 2012-02-28T01:02:21Z
dc.date.issued 2012-02-28T01:02:21Z
dc.identifier.uri http://jhir.library.jhu.edu/handle/1774.2/35726
dc.description.abstract An important task for computer vision systems is to segment adjacent structures in images without producing gaps or overlaps. Multi-object Level Set Methods (MLSM) perform this task with the benefit of sub-pixel accuracy. However, current implementations of MLSM are not as computationally or memory efficient as their region growing and graph cut counterparts which lack sub-pixel accuracy. To address this performance gap, we present a novel parallel implementation of MLSM that leverages the sparse properties of the segmentation algorithm to minimize its memory footprint for multiple objects. The new method, Multi-Object Geodesic Active Contours (MOGAC), can represent N objects with just two functions: a label image and unsigned distance field. The time complexity of the algorithm is shown to be O((M^d)/P) for M^d pixels and P processing units in dimension d={2,3}, independent of the number of objects. Results are presented for 2D and 3D image segmentation problems. en
dc.description.provenance Submitted by Michael Kazhdan (misha@cs.jhu.edu) on 2012-02-28T00:56:50Z No. of bitstreams: 1 mogac_technical_report.pdf: 698635 bytes, checksum: bd2e2ca338c6cc7b4ccce703bb8ee104 (MD5) en
dc.description.provenance Approved for entry into archive by Laura Graham(lgraham@jhu.edu) on 2012-02-28T01:02:21Z (GMT) No. of bitstreams: 1 mogac_technical_report.pdf: 698635 bytes, checksum: bd2e2ca338c6cc7b4ccce703bb8ee104 (MD5) en
dc.description.provenance Made available in DSpace on 2012-02-28T01:02:21Z (GMT). No. of bitstreams: 1 mogac_technical_report.pdf: 698635 bytes, checksum: bd2e2ca338c6cc7b4ccce703bb8ee104 (MD5) en
dc.language.iso en_US en
dc.relation.ispartofseries Johns Hopkins University Department of Computer Science;Technical Report 12-01
dc.title Multi-Object Geodesic Active Contours (MOGAC): A Parallel Sparse-Field Algorithm for Image Segmentation en
dc.type Article en

Files in this item

Files Size Format Download
mogac_technical_report.pdf 698.6Kb application/pdf Download

This item appears in the following Collection(s)

Show simple item record