摘要
Sankar K.Pal等最近提出了一种基于"粗糙熵"的图像分割算法,主要是按照目标和背景这两大类对图像进行分割,不足之处在于不能满足多类目标提取的需要。为此,基于商空间的粒度分解和粒度合成原理,综合粗糙集和聚类算法对之进行改进。通过对遥感图像进行分割处理,证明了改进后算法的有效性。
An image segmentation algorithm based on ＂rough entropy＂ is proposed by Sankar K. Pal,etc. In recendy, mainly according to two classes：object and background to segment an image. Defect of the algorithm is unsatisfactory for multi - class segmentation problem. Therefore, rough set and cluster methods are integrated based on granularity decompose and granularity synthesis theory of quotient space to improve this algorithm. The experimental result of remote sensing image segmentation demonstrate that the improved approach is valid.
出处
《计算机技术与发展》
2007年第9期58-60,共3页
Computer Technology and Development
基金
国家自然科学基金项目(60273043)
安徽省高校拔尖人才基金项目(05025102)
安徽省自然科学基金项目(050420204)
关键词
商空间
粗糙集
聚类
遥感图像分割
quotient space
rough set
cluster
remote sensing image segmentation
