摘要
颜色直方图作为颜色特征描述了图像颜色的统计分布特征,具有平移和旋转不变性,提出一种迭代自组织聚类分析算法可以用来实现基于颜色直方图的图像检索。该算法将ISODATA算法引入图像检索领域,实现了通过聚类算法来优化图像检索的目的。实验结果表明,该算法的应用大大提高了图像检索的速度以及图像命中率。
By means of improved Iterative Self-Organizing Data Analysis Techniques Algorithm, a new algorithm for color-histogram-based image retrieval has been proposed in this paper. The color histogram, as the color character, has been used to describe the statistic distributing character of the image color which has the feature of invariability in parellel moving and diversion. By introducing ISODATA into the field of image search, this algorithm can realize the purpose of optimizing, the image retrieval by the cluster algorithm. Experiment results show that the application of the algorithm has much improved the speed and the image-shooting rate of the image retrieval.
出处
《安徽工业大学学报（自然科学版）》
CAS
2006年第4期440-443,共4页
Journal of Anhui University of Technology（Natural Science）
关键词
图像检索
聚类
颜色直方图
image retrieval
cluster
color histogram
