Quaternionic wavelets for texture classification

摘要: This article proposes a study of the recent quaternionic wavelet transform (QWT) from practical point view through digital image analysis application. Based on theoretic 2D generalization analytic signal leading to strong modeling, this representation uses actual wavelets and yields subbands having shift-invariant magnitude 3-angle phase, using quaternion algebra. Our experiment furthers understanding quite sophisticated tool, shows its interest by clear improvement famous application: texture classification. Thanks coherent multiscale brought QWT we obtain better classification results than with standard in similar process.