Feature Extraction and Dimensionality Reduction Using Radon and Fourier Transforms with Application to Face Recognition

This paper presents a pattern recognition framework for face recognition using Radon and Fourier transforms. The technique computes Radon projections in different orientations and captures the directional features of the face images. Further, the Fourier transform applied on Radon projections provides spatial frequency features of the facial images. Being the line integral, Radon transform improves the low frequency components, which are useful in face recognition. As the Radon transform of the zero mean signals is zero, the approach is immune to zero mean additive noise. The approach has the advantage of dimensionality reduction. The reduced memory requirement makes the approach suitable for travel safety, physical and virtual access. For classification, minimum distance classifier has been used. The effectiveness of the algorithm is compared with different existing approaches for face recognition using the FERET and the ORL databases. Experimental results show the superiority of the proposed method in term of recognition rate with some of the existing popular algorithms.