全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

基于数学规划条件的粘连颗粒图像鞍点搜寻

DOI: 10.11834/jig.20100305

Keywords: 数学规划,鞍点规划,颗粒图像,鞍点

Full-Text   Cite this paper   Add to My Lib

Abstract:

鞍点在图像粘连对象分割的过程中起着非常关键的作用,目前已有若干以分割为目的的鞍点搜索算法,但在其简捷性、鲁棒性等方面仍有改善的余地。提出一种基于数学规划条件的颗粒图像鞍点搜寻新方法。该方法首先根据颗粒图像3维地貌化后得到的像素空间分布信息,利用数学规划原理提出颗粒图像的鞍点规划条件,然后根据其条件进行鞍点检测。将提出的算法应用在真实粘连颗粒图像上,实验结果表明,该算法具有较高的准判率,为颗粒图像的分割以及相关后续处理奠定了理论分析基础。

References

[1]  ,Kostas Zotos Object-oriented programming in mathematics,Applied Mathematics and Computation,2007(20).
[2]  Mahato K,Kailas S,Kapoor S S,Fission barriers and shell corrections at the saddle point for mass ~ 200,Progress in Particle and Nuclear Physics,2007(16).
[3]  Aleksandra Keli\'c,Karl-Heinz Schmidt,Assessment of saddlepoint mass predictions for astrophysical applications,Physics Letters B,2006(5).
[4]  Frank Fueten,Jeffrey Mason,An artificial neural net assisted approach to editing edges in petrographic image collected with the rotating polarizer stage,Computers & Geosciences,2007(3).
[5]  尤育赛.于慧敏.刘圆圆 基于粒度测量的重叠圆形颗粒图像分离方法 [J].-浙江大学学报(工学版)2005(7)
[6]  Vincent Tariel,Dominique Jeulin,Alain Fanget,3D multiscale segmentation of granular materials,Image Anal Storeol,2008(5).
[7]  Chen Ken,Saddle-point based separation of touched objects in 2-D image,Journal of Electronics,2006(3).
[8]  Chen Ken,Wang Yicong,Yang Rener,Hessian matrix based saddle point detection for granules segmentation in 2D image,2008(6).
[9]  Wu Zhiyuo,Zhang Liangsheng,The global optimization for some classes of programming problem,OR Transations,2003(2).
[10]  Lin Yiqin,Wei Yimin,A convergence analysis of the nonlinear Uzawa algorithm for saddle point problems,Applied Mathematics Letters,2007(4).
[11]  Olsen R A,Kroes G J,Comparison of methods for finding saddle points without knowledge of the final states,Journal of Chemical Physics,2004(2).
[12]  Kuijper A,On detecting all saddle points in 2D images,Pattern Recognition Letters,2004(9).

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133