Searcharxiv⌕ Search

arXiv subjects

Fernando González del Cueto

Publications and source records attributed to Fernando González del Cueto.

2 recordsLinked to original sources

Hyperspectral imaging from diffractograms

We consider an optical system with a periodic point spread function, such as a camera with a periodic diffractive filter array placed near the sensor. Such systems produce a monochrome image that we call ``diffractogram'', and that multiplexes the spatial and spectral information about the scene. We present a method to efficiently invert this forward model and recover the hyperspectral scene. The inverse problem is formulated as a large scale $\ell_1$-penalized least squares problem that is solved iteratively with the Alternating Direction Method of Multipliers (ADMM). A Krylov subspace solver, such as LSQR, may require tens to hundreds of iterations to approximatively solve the augmented least-squares problem required in each ADMM iteration. Here we present an exact direct solver for the augmented least-squares problems that exploits the periodicity of the point spread function. Each direct solve costs roughly as much as three LSQR iterations, eliminating inner iterations entirely. We validate the method on both synthetic and experimental datasets.

math.NA↗

Fast circular convolution with periodic kernels

We study the periodic circular convolution or PCC, where the convolution kernel depends periodically on the spatial location. We introduce two fast algorithms to compute the PCC that rely on the Discrete Fourier Transform. One of the algorithms reveals that the matrix representation of the PCC is unitarily equivalent to a block diagonal matrix, which facilitates the analysis of PCCs and operations such as their product and inversion. We present two sample applications. One to deconvolve the action of an optical system whose point spread function depends periodically in space. The second is to compute dispersion relations for a periodic chain of masses.

math.NA↗