A Numerically Stable Circular Harmonic Reconstruction Algorithm
Bessel functionscircular harmonic decompositioncomputerized tomographyFourier coefficientsFourier series in special orthogonal functionsharmonic analysisimage processingnumerical examplespositron CTRadon inversion algorithmsRadon transformrecurrence relationsx-ray CTZernike polynomials
Orthogonal polynomials and functions of hypergeometric type (Jacobi, Laguerre, Hermite, Askey scheme, etc.) (33C45) Fourier coefficients, Fourier series of functions with special properties, special Fourier series (42A16) Special integral transforms (Legendre, Hilbert, etc.) (44A15) Integral equations of the convolution type (Abel, Picard, Toeplitz and Wiener-Hopf type) (45E10) Numerical methods for integral transforms (65R10) Numerical methods for integral equations (65R20)
Let f be the 2D-Radon transform of g, i.e. f(p,\(\phi)\) is the integral of g over the line with perpendicular distance p to the origin whose normal makes an angle \(\phi\) with the x-axes. The paper studies the numerical computation of g from f based on circular harmonic decomposition: If \(g_ n\), \(f_ n\) are the Fourier coefficients of g, f, resp., then \[ \int^{\infty}_{0}f_{2n}(p)\cos (pt)dp=(-1)^ n\pi \int^{\infty}_{0}g_{2n}J_{2n}(rt)rdr, \] \(J_{2n}\) the first kind Bessel functions of order 2n. A similar formula holds for odd indices. These integral equations are solved using Zernike polynomials. Numerical examples are given.
This page was built for publication: A Numerically Stable Circular Harmonic Reconstruction Algorithm
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4725694)