کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
530850 869793 2012 21 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Interactive image segmentation by matching attributed relational graphs
موضوعات مرتبط
مهندسی و علوم پایه مهندسی کامپیوتر چشم انداز کامپیوتر و تشخیص الگو
پیش نمایش صفحه اول مقاله
Interactive image segmentation by matching attributed relational graphs
چکیده انگلیسی

A model-based graph matching approach is proposed for interactive image segmentation. It starts from an over-segmentation of the input image, exploiting color and spatial information among regions to propagate the labels from the regions marked by the user-provided seeds to the entire image. The region merging procedure is performed by matching two graphs: the input graph, representing the entire image; and the model graph, representing only the marked regions. The optimization is based on discrete search using deformed graphs to efficiently evaluate the spatial information. Note that by using a model-based approach, different interactive segmentation problems can be tackled: binary and multi-label segmentation of single images as well as of multiple similar images. Successful results for all these cases are presented, in addition to a comparison between our binary segmentation results and those obtained with state-of-the-art approaches. An implementation is available at http://structuralsegm.sourceforge.net/.


► It starts from an oversegmentation to propagate the labels by region merging.
► It is trivially extensible from binary to multi-label segmentation.
► Spatial relations are exploited besides color information.
► It can be combined with the connectivity constraint to reduce the user effort.
► By including a boundary refinement step, it produced state-of-the-art results.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Pattern Recognition - Volume 45, Issue 3, March 2012, Pages 1159–1179
نویسندگان
, , , , ,