融合空间及邻域信息的色彩直方图检索方法  

Color histogram image retrieval based on spatial and neighboring information

作　　者：杨珺[1] 王继成[1] 刘然[2] 

机构地区：[1]同济大学计算机系,上海201804 [2]重庆大学计算机学院,重庆400044

出　　处：《计算机工程与应用》2007年第27期158-160,共3页Computer Engineering and Applications

摘　　要：针对直方图无法反映色彩分布的空间信息这一缺陷,提出一种融合了空间及邻域信息的色彩直方图检索方法。以同一色彩构成的最大连通区域及其边缘区域的颜色粗糙度来反映同一颜色像素形成的区域特点及邻域信息,从而降低检索结果对色彩的绝对依赖,提高了检索的准确性。In order to overcome the defect that the conventional color histogram retrieval method is prone to lose the spatial information of colors,this paper presents a color histogram image retrieval method based on spatial and neighboring information. We use the biggest prominent region＇s area in one color and its neighboring area＇s color roughness to capture the regional traits and its neighboring information.By doing that,the absolute dependence on color can be reduced and the accuracy of the image retrieval can be improved.

关 键 词：基于内容的图像检索 色彩直方图 空间信息 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]