颜色密度直方图检索方法  

Color Desentiy Histogram Retrieval Method

作　　者：王小玲 毛宏燕 WANG Xiao-ling;MAO Hong-yan(School of Information Management,Shanghai Lixin University of Accounting and Finance,Shanghai 201620,China;Shanghai Key Libratory of Worthy Computing,East China Normal University,Shanghai 200062,China)

机构地区：[1]上海立信会计金融学院信息管理学院,上海201620 [2]华东师范大学上海高可信计算重点实验室,上海200062

出　　处：《小型微型计算机系统》2021年第10期2085-2088,共4页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(19JZD023)资助。

摘　　要：颜色是基于内容的图像检索的重要特征.传统颜色直方图由于只考虑色彩总量而无法区别色彩空间分布差异.本文提出了一种新的颜色密度直方图(Color Density Histogram CDH).通过计算主要颜色的密度,反映颜色的空间分布离散程度.密度大,颜色分布较集中,密度小,则颜色分布较分散.在HSV颜色空间上,使用CAREL的1000图像做测试数据集,在平均查准率、查全率上都优于颜色直方图方法.结果表明,CDH能够结合颜色和空间特征,比传统的颜色直方图具有具有更好的检索效果.Color is an important feature of content-based image retrieval.The traditional color histogram cannot the difference of color spatial distribution because it only considers the total amount of color.In this paper,a newColor Density Histogram(CDH) is proposed.By calculating the density of the main color,the spatial dispersion degree of the color is reflected.High density means concentrated color distribution,while lowdensity means scattered color distribution.In HSV color space,1000 images of CAREL were used as test data set,which were superior to color histogram method in average precision and recall.The results showthat CDH can combine color and spatial characteristics and has better retrieval effect than traditional color histogram.

关 键 词：基于内容的图像检索 颜色直方图 空间分布 基于颜色的图像检索 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]