摘要
提出了组合颜色和纹理特征的图像检索方法,通过对颜色进行量化然后提取颜色特征,计算图像之间的颜色距离。用Gabor变换提取图像的纹理特征然后根据L2距离计算图像的纹理距离,利用颜色特征和纹理特征的加权来求检索图像和查询图像的相似度,根据相似度的大小来进行图像的检索。实验结果表明,基于组合特征的图像检索方法优于单纯的图像检索方法。
A new method combining color feature and texture feature is presented. The first step color feature is extracted, next the distance is computed. To texture feature, L2 is used to compute the distance, At last, combine the distance both color and texture, according the distance the similarity is analyzed. It is demonstrated that the synergy resulting from the combination of color and texture is superior than using just color and texture.
出处
《计算机工程与设计》
CSCD
北大核心
2007年第17期4180-4182,共3页
Computer Engineering and Design
基金
国家自然科学基金项目(90612003)
关键词
图像检索
颜色特征提取
纹理特征提取
相似性计算
加权距离计算
image retrieval
color feature extracted
texture feature extracted
similarity compute
add weight distance computer
