摘要
在经典粗糙集分类模型的基础上利用变精度粗糙集模型,引入近似区分矩阵的概念,提出了一种基于变精度粗糙集的图像分类模型及其分类算法,在变精度粗糙集分类模型的基础上利用贝叶斯粗糙集模型,通过引入全局相对增益函数给出了贝叶斯粗糙集属性约简的另外一种算法,最后提出了一种基于贝叶斯粗糙集的图像分类模型及其分类算法。实验结果表明在处理决策表不协调的图像分类问题,贝叶斯粗糙集图像分类方法性能良好,分类准确和高效。
In this paper, the classical rough set classification model based on the use of variable precision rough set model, is similar to the distinction matrix which is proposed based on variable precision rough set model of image classification and classification algorithms. In the classification of variable precision rough set model based on the use of Bayesian rough set model, through the introduction of the overall function of the relative gain, Bayesian rough set attribute reduction algorithm is given for another, which finally proposes a rough set based on the Bayesian rough model and the image classification model and its classification algorithm. The experimental results show that in dealing with decision-making table which lack of coordination of the image classification problem, Bayesian image classification method of rough set is good condition, accurate and efficient classification.
出处
《信息技术》
2009年第9期129-131,共3页
Information Technology
关键词
变精度粗糙集
近似区分矩阵
贝叶斯粗糙集
全局相对增益
图像分类
variable precision rough set
approximate distinction matrix
Bayesian rough set model
overall relative gain
image classification
