基于Log-Polar和DT-CWT的旋转不变纹理分类算法

Rotation invariant texture classification algorithm based on Log-Polar and DT-CWT

提出了一种基于对数-极坐标变换(Log-Polar)和双树复数小波变换(DT-CWT)的旋转不变纹理分类算法。该方法首先对纹理图像进行对数-极坐标变换将旋转转化为平移,再用具有平移不变性的双树复数小波对变换后的图像滤波并计算各子带的能量值组成旋转不变特征向量,最后利用支持向量机算法实现纹理图像的分类。将该方法与其它旋转不变纹理分类算法进行比较,实验结果表明,提出的算法能有效地提高正确分类率。

A rotation invariant texture classification algorithm based on log-polar and Dual-Tree Complex Wavelet Transform（DT- CWT） is proposed.First,the texture image is translated by log-polar transform to convert the rotation to translation,the rotation invariant feature vector is composed of the energies of the subbands acquired by DT-CWT which is shift invariant to the translated texture image,the SVM algorithm is used to the texture classification at last.This method is compared with other rotation invariant texture classification algorithm,the experiment results show that it can improve the classification rate effectively.