Fig. 2: LSOM limits and experimental implementation.
a, Fundamental bounds on LSOM performance. Optimal resolution as a function of the FOV size D and the total number of scattered photons, with representative curves for 103, 106, 109 and 1012 photons. The FOV used in the experiments (D = 0.8λ) is indicated. b, Simplified schematic of the experimental LSOM set-up. An object is illuminated using TIR, and the scattered light is collected by a microscope. Slepian–Pollak coefficients at the Fourier plane are measured using a common-path interferometric single-pixel detection scheme with a reconfigurable mask. c, Sequential measurement of coefficients. A vector of Slepian–Pollak coefficients is obtained after N′ measurements, where the kth measurement is repeated Rkk times (k = 1, …, N′). A mask realizing the complex conjugate (denoted by *) of each Slepian–Pollak function, Sk (k = 1, … , N′), collimates the corresponding mode.
