基于多特征融合的面料图像检索系统

Fabric Image Retrieval System Based on Multi-feature Fusion

面料图像具有数据量多、信息量大的特点,传统的检索方法性能较低,不能实现对面料图像快速准确的检索。文章提出了一种基于多特征融合的检索方法,通过基于HSV颜色矩的算法提取颜色特征,采用Hu不变矩算法提取形状特征,利用Gabor小波提取纹理特征,将三种特征值设置不同的比重融合,整体比重值为1。根据相似性度量,对颜色特征、形状特征、纹理特征以及三项融合的特征进行图像检索,实验结果表明,采用基于多特征融合的方法进行面料图像检索,可以获得较好的检索效果。

The fabric image has the characteristics of large amount of data and large amount of information.The traditional re⁃trieval method has low performance and cannot realize fast and accurate retrieval of fabric images.In this paper,a multi-feature fu⁃sion based retrieval method is proposed.The color feature is extracted by the HSV color moment algorithm.The Hu invariant mo⁃ment algorithm is used to extract the shape features.The Gabor wavelet is used to extract the texture features,and the three feature values are set to different weights.Fusion,the overall weight value is 1.According to the similarity measure,the image retrieval of color features,shape features,texture features and three fusion features is carried out.Experimental results show that the use of multi-feature fusion method for fabric image retrieval can achieve good retrieval results.