摘要
相容粒空间模型是粒计算理论的主要方法之一,利用粒计算的方法对相容粒及其空间模型的研究已经比较深入,然而在实际图像分类的应用中,面对的问题更为复杂。为了解决现实图像分类中存在的信息有限或信息缺失而造成分类不准确的问题,本文重点研究相容粒模型的分层思想,通过与善于解决以上问题的三支决策理论相结合,形成一种基于相容粒模型和三支决策理论相结合的图像分类算法,并通过仿真实验验证该算法是有效可行的。
Tolerance Granular Model is one of the main methods of granular computing theory, using the method and its model of granular computing to tolerance granular has been deeply studied, however in the practical application of image classification, we face the problem more complicated. In order to solve the inaccurate problems of the realistic image classifi- cation which caused by the information limited or lack, this paper focuses on the research of the layered idea of tolerance granular model, combined with delay decision in the three-way decisions which is good at the above problems, form a kind of image classification algorithm based on tolerance granular model and three-way decisions theory, and the effectiveness of the algorithm is verified by experiment.
出处
《计算技术与自动化》
2014年第4期93-96,共4页
Computing Technology and Automation
基金
国家自然科学基金项目(60873104
61040037)
河南省科技攻关重点计划项目(112102210194)
关键词
相容粒
三支决策
图像分类
tolerance granular
Three-way decisions
image classification
