一种多特征融合的外观设计专利图像检索方法  

Image retrieval method for design patent based on multi-feature fusion

作　　者：张国宏[1] 蔡念[1] 楼朋旭[1] 戴青云[1] 

机构地区：[1]广东工业大学信息工程学院,广州510006

出　　处：《计算机工程与应用》2011年第14期166-169,共4页Computer Engineering and Applications

基　　金：广东省产学研项目(No.2008B090500254)

摘　　要：提出一种多特征融合的外观设计专利图像检索方法。采用Hu不变矩和高斯描绘子两种算法描述图像形状特征;采用加权法融合形状特征和Gabor小波提取的纹理特征;用Adaboost算法进行相关性反馈。实验结果表明,提出的方法可以有效地实现外观设计专利图像的检索。An image retrieval method for design patent is proposed based on multi-feature fusion.Invariant moments and Gaussian descriptor describe shape features of design patent images.Shape features and texture features obtained by using Ga- bor wavelet are integrated with a weighted method.Adaboost algorithm is employed for relevance feedback.The experimental results indicate that the proposed method is feasible for design patent image retrieval.

关 键 词：外观设计专利 图像检索 高斯描绘子 多特征融合 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]