The goal of this study is to compare the effect of rotation variance on dual-tree complex wavelet transform (DT-CWT) based tissue characterization methods. Rotation variant features are especially useful when estimating orientations of textures. On the other hand, since tissues can turn and orient themselves on different directions, in segmentation or classification problems, rotation invariance becomes important. The methods are tested on two texture compositions from the Brodatz texture database and two actual magnetic resonance (MR) images. Results show that, the classification performance is not significantly affected by using rotation variant or invariant methods. On the other hand, regarding segmentation, rotation invariant DT-CWT features performed slightly better compared to rotation variant features.
Add the full text or supplementary notes for the publication here using Markdown formatting.