Image Reconstruction by Chebyshev-Fourier Moments

In this research, we have analyzed the Chebyshev-Fourier moments computing and found out that the errors are mainly caused by the computational issues of both radial and Fourier polynomials. To increase the accuracy of moments computing, we utilized the k × k numerical scheme in the computations of Chebyshev-Fourier moments and conducted the image reconstructions to verify our solution. The experimental results show that the performances of image reconstructions are highly satisfied. We have also performed image reconstructions from the different radial and Fourier orders of moments, and observed that the radial and Fourier orders of Chebyshev-Fourier moments preserve the information of circular and radial patterns, respectively.