| 翻訳と辞書 |
| grassfire transform ： ウィキペディア英語版 |
grassfire transform
In image processing, the grassfire transform is the computation of the distance from a pixel to the border of a region. It can be described as "setting fire" to the borders of an image region to yield descriptors such as the region's skeleton or medial axis. Harry Blum introduced the concept in 1967.〔Blum, Harry. ''A Transformation for extracting new descriptors of shape'', 1967,"http://pageperso.lif.univ-mrs.fr/~edouard.thiel/rech/1967-blum.pdf",6/8/2012〕
==Motivation==
A region's skeleton can be a useful descriptor, because it describes things such as the symmetry of the region as well as subparts, depressions and protrusions.〔Leymarie, F. and Levine, M.,''Simulating the Grassfire Transform Using an Active contour Model'',IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.14, NO 1, January 1992,"http://portal.acm.org/citation.cfm?id=123438"〕 It also provides a way of relating the interior of a region to the shape of the boundary. In the grassfire transform, the skeleton forms at the points in the region where the "fires" meet. In the literature this is described as the locus of meeting waveforms.〔
Another advantage of using the outcome of the grassfire transform as a descriptor is that it is invertible. Assuming information about when the medial axis or skeleton is created by meeting waveforms is kept, then the skeleton can be restored by radiating outward.〔
抄文引用元・出典: フリー百科事典『 ウィキペディア（Wikipedia）』
■ウィキペディアで「grassfire transform」の詳細全文を読む
スポンサード リンク
| 翻訳と辞書 : 翻訳のためのインターネットリソース |
Copyright(C) kotoba.ne.jp 1997-2016. All Rights Reserved.
|
|