| 基于MPEG-7多特征融合的图像检索技术 |
| |
| 引用本文： | 樊凌. 基于MPEG-7多特征融合的图像检索技术[J]. 电脑与信息技术, 2011, 19(6): 12-15 |
| |
| 作者姓名： | 樊凌 |
| |
| 作者单位： | 苏州市职业大学教务处,江苏苏州,215104 |
| |
| 摘 要： | 针对MPEG-7单一描述符特征检索效率低的问题，提出了主颜色和边缘直方图特征融合的检索算法。同时利用检索目标在图像中分布的特点，利用分块加权法来提高对检索主目标的检索准确度。实验表明，该方法在查准率和查全率均优于单一特征的检索算法，检索精度得到了提高。
|
| 关 键 词： | MPEG-7 图像检索 HSV直方图：特征融合：相似度度量 |
Image Retrieval Technology Based on Multi-Feature Fusion in MPEG-7 |
| |
| FAN Ling. Image Retrieval Technology Based on Multi-Feature Fusion in MPEG-7[J]. Computer and Information Technology, 2011, 19(6): 12-15 |
| |
| Authors: | FAN Ling |
| |
| Affiliation: | FAN Ling(Office of Academic Affairs,Suzhou Vocational University,Suzhou 215104,China) |
| |
| Abstract: | For MPEG-7 low efficiency problems in image retrieval by using a single feature retrieval , the paper presents a new content-based image retrieval method on main color and edge histogram feature fusion. Retrieval accuracy can be improved by retrieval target characteristics of the distribution and block-weighted. Experiments show that the method in the precision rate and recall rate is better than a single feature of the retrieval algorithms, retrieval accuracy is improved. |
| |
| Keywords: | MPEG-7 image retrieval HSV histogram similarity metrics feature fusion |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
