Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (14): 166-169.

• 图形、图像、模式识别 • Previous Articles     Next Articles

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

ZHANG Guohong,CAI Nian,LOU Pengxu,DAI Qingyun   

  1. School of Information Engineering,Guangdong University of Technology,Guangzhou 510006,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-05-11 Published:2011-05-11

一种多特征融合的外观设计专利图像检索方法

张国宏,蔡 念,楼朋旭,戴青云   

  1. 广东工业大学 信息工程学院,广州 510006

Abstract: 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 Gabor 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.

Key words: design patent, image retrieval, Gaussian descriptor, multi-feature fusion

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。

关键词: 外观设计专利, 图像检索, 高斯描绘子, 多特征融合