论文部分内容阅读
基于内容的图像检索是数字图书馆和多媒体数据库的重要研究内容。文中主要针对图像处理，特别是运用图像的纹理特性来查找或检索图像数据库。运用Ｇａｂｏｒ 变换进行纹理分析是比较成功的方法。小波变换尽管在这方面也取得了较好的效果，但在许多情况下，不如用Ｇａｂｏｒ滤波器效果好。文中就是在Ｇａｂｏｒ 滤波器的基础上生成Ｇａｂｏｒ 小波，对图像进行多分辨率分析，提取纹理特征，提供了一种基于图像纹理的图像检索方法，并对此给出了一些实验结果。
Content-based image retrieval is an important research content of digital library and multimedia database. This paper focuses on image processing, especially the use of image texture features to find or retrieve the image database. The use of Gabor transform texture analysis is a more successful method. Although wavelet transform has achieved good results in this aspect, it is better than Gabor filter in many cases. In this paper, Gabor wavelet is generated on the basis of Gabor filter, and the multi-resolution analysis of the image is carried out to extract the texture features. An image retrieval method based on image texture is provided, and some experimental results are given.