|国家预印本平台
首页|基于Gabor小波变换的纹理图像检索技术研究

基于Gabor小波变换的纹理图像检索技术研究

Research on Texture Images Retrieval Based on the Gabor Wavelet Transform

中文摘要英文摘要

本文采用基于Gabor小波变换的纹理图像检索方法,以Brodatz纹理库为对象进行了实验。以Gabor函数为母波函数生成一组不同方向不同尺度的滤波器,对纹理图像进行小波变换后,把变换系数的均值和方差作为图像的纹理特征,然后求绝对距离进行检索。实验表明,这种方法对纹理图像具有优良的检索性能。

technique of texture images retrieval based on Gabor wavelet transform is presented in this paper.A group of filters of different directions and different scales, generated from the basic wavelet of Gabor function are used for the wavelet transforamtion of texture images,then the means and Variances of the wavelet transformation coefficients are calculated as texture features of the image for the use of image retrieval. The parameters used in experiments ars circumspectly selected according to the retrieval results.Experiments show that this method has an excellent performance on texture images retrieval.

杨杰、周亮

电子技术应用

Gabor小波纹理相似性图像检索

Gabor waveletTexturesimilarityImage Retrieval

杨杰,周亮.基于Gabor小波变换的纹理图像检索技术研究[EB/OL].(2008-06-18)[2025-08-11].http://www.paper.edu.cn/releasepaper/content/200806-427.点此复制

评论