A skeleton pruning algorithm based on information fusion

A skeleton pruning algorithm based on information fusion
复制标题

一种基于信息融合的骨架剪枝算法

DOI:
10.1016/j.patrec.2013.03.013
复制
发表时间:
2013-07
影响因子:
5.1
通讯作者:
Hsu D. Frank
Hsu D. Frank
中科院分区:
计算机科学3区
文献类型:
--
作者:
Liu Hongzhi;Wu Zhonghai;Zhang Xing;Hsu D. Frank

文献摘要

参考文献

被引文献

相似文献

骨架剪枝是骨架处理和分析的重要组成部分。由于缺乏对分支重要性或意义的标准测量,这仍然是一个相当具有挑战性的问题。同一分支,如果从不同的角度、不同的目标来看,其相对意义也会不同。不同的客观测量有其优点和局限性。为了整合不同客观测量的优点,我们将骨架剪枝视为多目标决策问题,并提出一种基于信息融合的骨架剪枝算法。在剪枝过程中,我们使用组合融合分析和认知多样性的概念来融合分支重要性的各种测量,包括区域重建、轮廓重建和视觉贡献。实验结果表明:(1)该方法在各种形状上都是稳定的,并且对边界噪声具有鲁棒性;(2)它可以根据视觉判断有效地生成多尺度骨架。
Skeleton pruning is an essential part of the processing and analysis of skeletons. It is still quite a challenging problem because of the lack of standard measurements for the importance or significance of a branch. The relative significance of the same branches will be different if we see them from different perspectives with different objectives. Different objective measurements have their advantages and limitations. To integrate the advantages of different objective measurements, we consider skeleton pruning as a multi-objective decision-making problem and propose a skeleton pruning algorithm based on information fusion. During the pruning process, we use combinatorial fusion analysis and the concept of cognitive diversity to fuse various measurements of branch significance including region reconstruction, contour reconstruction and visual contribution. Experimental results show that: (1) the proposed method is stable across a wide range of shapes and robust to boundary noise, and (2) it can effectively generate multi-scale skeletons according with visual judgment.
DOI: 10.1016/0020-0255(91)90020-u
发表时间: 1991-08
期刊: Inf. Sci.
影响因子: --
作者:
S. Yau;Ivy Wiharja
通讯作者: S. Yau;Ivy Wiharja
骨架剪枝是骨架简单性和重建误差之间的权衡
DOI: 10.1007/s11432-012-4715-3
发表时间: 2013-01
期刊: Science China Information Sciences
影响因子: --
作者:
Wei Shen;Xiang Bai;Xingwei Yang;Longin Latecki
通讯作者: Longin Latecki
DOI: 10.1016/j.patcog.2010.08.021
发表时间: 2011-02
期刊: Pattern Recognit.
影响因子: --
作者:
W. Shen;X. Bai;Rong Hu;Hongyuan Wang;Longin Jan Latecki
通讯作者: W. Shen;X. Bai;Rong Hu;Hongyuan Wang;Longin Jan Latecki
DOI: 10.1016/j.patrec.2012.07.014
发表时间: 2012-12
期刊: Pattern Recognit. Lett.
影响因子: --
作者:
Hongzhi Liu;Zhonghai Wu;D. Hsu;B. Peterson;Dongrong Xu
通讯作者: Hongzhi Liu;Zhonghai Wu;D. Hsu;B. Peterson;Dongrong Xu
DOI: 10.1109/icip.1996.560357
发表时间: 1996-09
期刊: Proceedings of 3rd IEEE International Conference on Image Processing
影响因子: --
作者:
D. Attali;A. Montanvert
通讯作者: D. Attali;A. Montanvert