Rotational-invariant texture analysis using Radon and Polar complex exponential transform

Abstract

Rotational invariant texture analysis technique using Radon and PCET is proposed in this stab. Rotation invariance is achieved within the Radon space. Translation invariance is achieved by normalization of moment of PCET. A k-nearest neighbor classifier is employed for classifying the texture. To test and evaluate the proposed method several sets of texture was evaluated. The evaluation is achieved with different scaling, translation and rotation under different noisy conditions. Correct classification percentage is calculated for various noise conditions. Experimental result shows pre-eminence of the employed method as compared to the recent invariant texture analysis methods.