ORTHOGONAL FOURIER-MELLIN MOMENTS FOR INVARIANT PATTERN-RECOGNITION

We propose orthogonal Fourier-Mellin moments, which are more suitable than Zernike moments, for scale-and rotation-invariant pattern recogni tion. The new orthogonal radial polynomials have more zeros than do th e Zernike radial polynomials in the region of small radial distance. T he orthogonal Fourier-Mellin moments may be thought of as generalized Zernike moments and orthogonalized complex moments. For small images, the description by the orthogonal Fourier-Mellin moments is better tha n that by the Zernike moments in terms of image-reconstruction errors and signal-to-noise ratio. Experimental results are shown.