| 基于颜色、形状和纹理的多特征融合图像检索 |
| |
| 引用本文： | 李薇,李秉茂.基于颜色、形状和纹理的多特征融合图像检索[J].航空计算技术,2013(6):88-90,107. |
| |
| 作者姓名： | 李薇 李秉茂 |
| |
| 作者单位： | [1]中国航空综合技术研究所,北京100028 [2]山西建筑职业技术学院计算机工程系,山西太原030006 |
| |
| 摘 要： | 基于单一特征的图像检索的许多方法都已经相当成熟，如何有效利用这些特征是基于内容图像检索的一个关键问题。利用颜色、纹理、形状等特征，在多特征融合检索时用动态生成权值的方法进行图像检索。实验结果表明，算法能够准确、高效地检索出目标图像，相对于单一特征的检索，有效提高了图像的检索精度。
|
| 关 键 词： | 多特征图像检索 动态权值 形状特征 纹理特征 |
Content- based Image Retrieval using Combination Features |
| |
| Institution: | LI Wei1 , LI Bing- mao2 ( 1. China Aero- Polytechnology Establishment, Beifing 100028, China ; 2. Computer and Engineering Department, Shanxi Architectural College, Taiyuan 030006, China) |
| |
| Abstract: | Image retrieval based on single feature is already quite mature, how to use these single features effectively is the key issue in improving the effect of the image retrieval. We often use the combination features to solve the problem. In this paper, we use the dynamic weight to combination these single fea- tures. Experimental results show that the algorithm using dynamic weight can accurately and efficiently re- trieve the target image compared with using the method of using fixed weight. The method using the dy- namic weight effectively improves the image retrieval accuracy. |
| |
| Keywords: | multi- features image retrieval dynamic weight fixed weight |
| 本文献已被 维普 等数据库收录！ |
|
