DocumentCode :
2382706
Title :
Countering oversegmentation in partitioning-based connectivities
Author :
Ouzounis, Georgios K. ; Wilkinson, Michael H F
Author_Institution :
Inst. for Math. & Comput. Sci., Groningen Univ., Netherlands
Volume :
3
fYear :
2005
fDate :
11-14 Sept. 2005
Abstract :
A new theoretical development is presented for handling the over-segmentation problem in partitioning-based connected openings. The definition we propose treats singletons generated with the earlier method, as elements of a larger connected component. Unlike the existing formalism, this new method allows detection of filamentous structures linking larger objects. This is demonstrated in attribute filtering on neuron images.
Keywords :
filtering theory; image segmentation; countering oversegmentation; filamentous structures; neuron images filtering; partitioning-based connected openings; partitioning-based connectivities; Filtering; Filters; Gray-scale; Joining processes; Mathematics; Morphology; Neurons; Object detection; Pixel; Robustness;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing, 2005. ICIP 2005. IEEE International Conference on
Print_ISBN :
0-7803-9134-9
Type :
conf
DOI :
10.1109/ICIP.2005.1530524
Filename :
1530524
Link To Document :
بازگشت