摘要
高维图像特征数据不利于图像数据挖掘。为了降低图像特征数据维数,提出了基于概念格的降维算法,该算法将图像的HSV颜色特征转换为图像形式背景,再对背景的概念格进行属性约简。实验结果表明,该降维方法比较有效,并且较主成分分析具有明显的优势。
High-dimensional image feature data is an obstacle for image mining. In order to reduce the image feature data dimension, this paper introduced a dimension reduction algorithm based on concept lattices. In this algorithm, calculated the concept lattices of the image formal, which was transformed from the HSV color feature, and then reduced the attributes of the concept lattice. The experiment results prove that this method is efficient, and it outperforms the principal component analysis （PCA）.
出处
《计算机应用研究》
CSCD
北大核心
2009年第9期3553-3555,共3页
Application Research of Computers
基金
国家自然科学基金资助项目(60775036)
辽宁省自然科学基金资助项目(20082189)
关键词
图像挖掘
维度灾难
形式背景
概念格
属性约简
数据降维
image mining
curse of dimensionality
formal context
concept lattices
attribute reduction
data dimension reduction
