摘要
提出综合纹理、颜色和形状特征的图像检索方法。首先采用Gabor小波计算ROIs(Regions of Interest)的位置和数目;然后在ROIs中,使用Gabor小波提取纹理特征,采用YUV空间直方图和颜色矩表示颜色特征,使用Zernike矩提取形状特征。为了提高图像检索的准确度,最后采用基于支持向量机(SVM)的相关反馈算法。实验结果表明,提出的方法具有较好的检索性能。
The main focus in this paper is on integrated color, texture and shape extraction methods for CBIR. Original CBIR methodology that uses Gabor filtration for determining the number of Regions of Interest （ROIs）. In the ROIs extracted, texture features based on thresholded Gabor features, color features based on histograms, color moments in YUV space, and shape features based on Zernike moments are then calculated. At last,an algorithm of the support vector machine for improving veracity of image retrieval is applied. The result of experiment illustrate proposed method have a better retrieval performance.
出处
《现代电子技术》
2008年第16期143-146,共4页
Modern Electronics Technique
基金
国家自然科学基金重点资助项目(60234030)
关键词
图像检索
特征提取
ZERNIKE矩
纹理分析
相关反馈
支持向量机
image retrieval
feature extraction
Zernike moments
texture analysis
relevance feedback
support vector machine
