Two-dimensional NUFFT-based algorithm for fast near-field imaging

A fast and accurate algorithm for near-field imaging is presented. First, a two-dimensional (2-D) nonuniform fast Fourier transformation (NUFFT) is directly applied to the near-field scattered data. Second, the circular wavefront curvature is compensated by mapping the NUFFT images to a new rectified grid. The 2-D NUFFT-based algorithm directly exploits the data collected over a polar grid and accordingly avoids the polar to rectangular interpolation. Through numerical simulations, we show that the 2-D NUFFT-based algorithm can provide accurate images with lower sidelobes and is much faster than the near-field tomographic method.