关键词: 旋转不变, 纹理分类, 对数-极坐标变换, 双树复数小波变换, 支持向量机

Abstract: 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.

Key words: rotation invariance, texture classification, log-polar transform, dual-tree complex wavelet transform, support vector machines