利用NSCT和Krawtchouk矩进行图像检索

Image Retrieval Based on NSCT and Krawtchouk Moments

摘要: 提出了一种基于非下采样Contourlet变换(nonsubsampled contourlet transform,NSCT)和Krawtchouk矩的图像检索算法。首先,通过NSCT对图像进行分解,提取每个分解层次上不同方向子带系数分布的数学特征作为图像的纹理特征;然后,利用Krawtchouk矩描述图像的形状特征;最后,根据加权的相似性度量实现图像检索。实验结果表明,所提取的特征向量具有平移、旋转、尺度不变性,且能获得更高的检索精度。

Abstract: To further improve the accuracy of image retrieval,an algorithm based on nonsubsampled contourlet transform(NSCT) and Krawtchouk moments is proposed.Firstly,the image is decomposed to the low resolution image with NSCT,extracting the contourlet coefficients of each decomposition level of different direction and using the statistical features of the coefficients as texture features of the image.Then,the Krawtchouk moments are employed to extract shape features of the image.Finally,the image retrieval is achieved based on weight distance measure.A comparison is made with the method of using NSCT and Zernike features.Experimental results show that the extracted features have the properties of translation,rotation and scaling invariance,and the proposed algorithm can acquire the higher retrieval accuracy.