当前位置: 首页 > 文章 > 基于多特征融合的图像语义标注 东北林业大学学报 2008,36 (10) 90-91
Position: Home > Articles > Image Semantic Labeling Based on Multiple Feature Fusion Journal of Northeast Forestry University 2008,36 (10) 90-91

基于多特征融合的图像语义标注

作  者:
胡全;邱兆文;王霓虹
单  位:
东北林业大学
关键词:
基于内容的图像检索;多特征融合;支持向量机;图像语义标注
摘  要:
采用了新的颜色特征提取方法,融合图像的颜色和纹理特征作为图像的特征向量,用支持向量机实现图像语义信息的标注。实验结果表明,多特征图像检索要比单一特征检索效果好,在颜色特征的基础上引入纹理特征和形状特征后可有效提高检索效率,而且采用支持向量机融合多特征可成功用于图像语义的标注。
译  名:
Image Semantic Labeling Based on Multiple Feature Fusion
作  者:
Hu Quan,Qiu Zhaowen,Wang Nihong(College of Information and Computer Engineering,Northeast Forestry University,Harbin 150040,P.R.China)
关键词:
Content-based image retrieval;Multiple feature fusion;Support vector machines;Image semantic labeling
摘  要:
A new color feature extraction method was applied to image semantic labeling with support vector machine by integrating color feature and texture feature into eigenvector.Experimental result shows that the retrieval of image with multiple features is better than that of image with single feature.The introduction of shape and texture features into color feature can improve the efficiency of image retrieval.Furthermore,support vector machine can be successfully used in labeling image semantic information.

相似文章

计量
文章访问数: 8
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊