Invariant Texture Recognition Using a Steerable Pyramid

A rotation-invariant texture recognition system is presented. A steerable oriented pyramid is used to extract representative features for the input textures. The steerability of the filter set allows a shift to a rotationally invariant representation via a DFT-encoding step. Supervised classification follows. State-of-the-art recognition results are presented for a 30 texture database with a comparison across the performance of the K-nn, Back-Propagation and Rule-Based classifiers. In addition, high accuracy estimation of the input rotation angle is demonstrated. The extension of the system to scale-invariance is discussed. We assess the performance of scale-invariant classification and we present a simple algorithm for doing so. 1 Introduction A number of texture recognition systems have been recently proposed in the literature, giving high-accuracy classification rates (e.g., [24, 14, 17, 13]) on textures whose orientation and scale were controlled. In object recognition and inspect..