| 摘要: |
| 图像的低层可视特征与高层语义特征之间存在着一道鸿沟,人们不能直接理解由计算机自动生成的低层特征.另外,基于内容的图像分类和检索的性能极大地依赖于可视特征的提取和描述.出于这些考虑,提出了新的图像纹理、边缘描述子提取方法,并将它们表示为直方图.在此基础上,集成纹理、边缘和颜色直方图作为图像的特征向量,用支持向量机(SVM)实现图像的语义分类.实验结果表明,集成的图像特征表示在图像分类实验中取得了很好的效果,具有比其他特征表示(如Gabor纹理、颜色直方图)更好的性能. |
| 关键词: 基于内容 图像特征描述子 颜色 纹理 边缘 分类 SVM |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Image Semantic Classification by Using SVM |
|
WAN Hua-Lin,Morshed U. Chowdhury
|
| Abstract: |
| There exists an enormous gap between low-level visual feature and high-level semantic information, and the accuracy of content-based image classification and retrieval depends greatly on the description of low-level visual features. Taking this into consideration, a novel texture and edge descriptor is proposed in this paper, which can be represented with a histogram. Furthermore, with the incorporation of the color, texture and edge histograms seamlessly, the images are grouped into semantic classes using a support vector machine (SVM). Experiment results show that the combination descriptor is more discriminative than other feature descriptors such as Gabor texture. |
| Key words: content-based image feature descriptor color texture edge classification SVM |
