Compressive parallel single-pixel imaging for efficient 3D shape measurement in the presence of strong interreflections by using a sampling Fourier strategy

We present a compressive parallel single-pixel imaging (cPSI) method, which applies compressive sensing in the context of PSI, to achieve highly efficient light transport coefficients capture and 3D reconstruction in the presence of strong interreflections. A characteristic-based sampling strategy is introduced that has sampling frequencies with high energy and high probability. The characteristic-based sampling strategy is compared with various state-of-the-art sampling strategies, including the square, circular, uniform random, and distance-based sampling strategies. Experimental results demonstrate that the characteristic-based sampling strategy exhibits the best performance, and cPSI can obtain highly accurate 3D shape data in the presence of strong interreflections with high efficiency.