点击次数：
论文类型：期刊论文
发表时间：2015-02-01
发表刊物：PATTERN RECOGNITION
收录刊物：SCIE、EI
卷号：48
期号：2
页面范围：402-410
ISSN号：0031-3203
关键字：Shape representation; Shape matching; Generalized electrical potential; Electrical charge distribution
摘要：Motivated by the fact that electrical charge distributions are almost the same for similar shapes but not vice versa when shapes reach their electrical equilibrium condition, we propose a novel shape signature based on the electrical charge distribution on the shape (ECDS). Compared to other shape signatures, ECDS has the following interesting properties: (1) ECDS is a local measure but computed in a global manner. Thus, it is more robust to noise and shape variations. (2) ECDS is articulation insensitive and therefore exhibits better performance by the introduction of generalized coulomb potentials. This allows it to better match shapes whose parts can move independently, such as scissors. (3) The sum of ECDS remains constant during the process of reaching electrical equilibrium, which does favor some applications. Numerous experiments have been done on several public shape databases (MPEG-7 database, articulated shape data set, Kimia silhouettes and ETH-80 data set), demonstrating that ECDS has the above properties and compares well with other shape descriptors in many kinds of shape retrieval and recognition tasks. (C) 2014 Elsevier Ltd. All rights reserved.