Transactions of the Society of Instrument and Control Engineers
Online ISSN : 1883-8189
Print ISSN : 0453-4654
ISSN-L : 0453-4654
Image Figure Segmentation Using Morphological Skeleton
Naoki TAKUMAKoichiro DEGUCHIIwao MORISHITA
Author information
JOURNAL FREE ACCESS

1993 Volume 29 Issue 11 Pages 1361-1368

Details
Abstract
A new method for image figure segmentation using morphological techniques is proposed. Morphological skeleton transform, which extract a medial axis of an input figure, is known to have an advantage that the original figure can be reconstructed from its resulted skeleton. At the same time, the skeleton is dissociated at constricted parts (or necks) of the input figure.
In this method, we utilize these two properties for segmentations. First, we take the morphological skeleton for the input image figure. Then, we reconstruct the original figure part by part separatedly from each of dissociated skeletons, and obtain segmented figures.
It will be a great advantage that, in this method, such a global image analysis as figure segmentation can be achieved only using steps of local morphological operations.
Content from these authors
© The Society of Instrument and Control Engineers (SICE)
Previous article Next article
feedback
Top