首页 | 本学科首页   官方微博 | 高级检索  
     

基于主色提取和主色集扩充的图像检索算法
引用本文:李丽丽,孙劲光. 基于主色提取和主色集扩充的图像检索算法[J]. 计算机应用与软件, 2012, 0(1): 152-154,167
作者姓名:李丽丽  孙劲光
作者单位:辽宁工程技术大学测绘与地理科学学院;辽宁工程技术大学电子与信息工程学院
基金项目:辽宁省教育厅基金(L2010158)
摘    要:提出一种基于主色提取和主色集扩充的图像检索算法。算法首先对图像进行主色特征提取,采用提出的改进主色集扩充方法分别对两图像进行主色集扩展,使其维数相同;然后,将经过主色提取得到的主色集从传统RGB颜色空间转换为具有良好视觉一致性HSV颜色空间;最后,采用Quadratic距离进行图像的相似性度量。实验证明算法易于实现、具有较好的时间复杂度及较高的查准率和查全率。

关 键 词:主色提取  主色集  图像检索  颜色空间  相似性度量

IMAGE RETRIEVAL ALGORITHM BASED ON MAIN COLOR EXTRACTION AND MAIN COLOR SET EXPANSION
Li Lili,Sun Jinguang. IMAGE RETRIEVAL ALGORITHM BASED ON MAIN COLOR EXTRACTION AND MAIN COLOR SET EXPANSION[J]. Computer Applications and Software, 2012, 0(1): 152-154,167
Authors:Li Lili  Sun Jinguang
Affiliation:1(School of Geomatics,Liaoning Technical University,Fuxin 123000,Liaoning,China) 2(School of Electronic and Information Engineering,Liaoning Technical University,Huludao 125105,Liaoning,China)
Abstract:The authors put forward an image retrieval algorithm based on main color extraction and main color set expansion.At the beginning,the algorithm extracts main color feature from images;then it uses an improved main color set expansion method,which is proposed in the article,to expand main color sets of two images one by one,until they are of the same dimension.Next,it transforms the main color sets,which are acquired by main color extraction,from the traditional RGB color space to HSV color space with good visual consistency.Finally,it uses Quadratic distance to measure image similarity.Experiments have shown that the algorithm is easy to implement while it bears better time complexity,higher precision rate and recall rate.
Keywords:Main color extraction Main color set Image retrieval Color space Similarity measure
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号