节点文献
基于语义的图像物体提取的新方法
A Novel Extracting Image Semantic Objects Based on Semantic
【摘要】 在图像语义研究中,提取图像中的语义物体或区域是重要的。本文首先对图像预处理,通过颜色空间的转换,在空间对图像进行K-均值分类,提取出具有语义性质的物体和区域。实验结果表明该方法是可行的,而且很有效的。
【Abstract】 The color classification process requires to partition a color image into uniform color regions.It is very important that ex-tract interesting region or object in image semantic analysis.In this paper we propose an approach to extract semantic region based on color feature.First,we translate RGB space into Lab space.Second,we use the k-means algorithm solve clustering problem.In the end,we extract semantic object in terms of color information.Experimental results show that the color clustering give superior re-sults in increases in cluster effectiveness.
【关键词】 彩色分类;
K-均值算法;
语义信息;
物体提取;
【Key words】 Color classification; K-means clustering; Semantic information; Object extraction;
【Key words】 Color classification; K-means clustering; Semantic information; Object extraction;
【基金】 国家自然科学基金(#60372068)资助;广东省自然科学基金(#06300098)资助
- 【文献出处】 微计算机信息 ,Microcomputer Information , 编辑部邮箱 ,2007年21期
- 【分类号】TP391.41
- 【被引频次】4
- 【下载频次】139