Fig. 1. Experimental schematic of the GI and SMGI systems with a pseudo-thermal light source. In the “object” arm, Z1 is the distance between the aperture and the object. In the “reference” arm, Z2 is the distance between the aperture and the reference CCD. The distance f is the focal length of lens L1.

Fig. 2. Calculated results for (a) ΦTΦ and (b) (ΦT−ΦX)Φ. The x axis represents the row coordinate of the matrices, and the y axis represents the column coordinate.

Fig. 3. Comparisons of experimental results obtained by GI, DGI, and SMGI for object “GI” for a varying number of measurements. (a) Reconstruction results corresponding to 500, 1000, and 2000 measurements, (b) PSNR for the GI (black), DGI (blue), and SMGI (red) reconstructions, plotted as a function of the number of measurements (100–2000).

Fig. 4. Calculated results for ΦTΦ. (a) Part of the diagonal region of ΦTΦ. (b) 2D grayscale image corresponding to (a). (c) Values of the matrix elements along the red dashed arrow lines in (a) and (b). The horizontal axis represents the sequence number of elements along the red dashed arrow line.

Fig. 5. Calculated results for (ΦT−ΦX)Φ. (a) Part of the diagonal region of (ΦT−ΦX)Φ; (b) 2D gray-scale image corresponding to (a); (c) Values of the matrix elements along the red dashed arrow lines in (a) and (b). The horizontal axis represents the sequence number of the elements along the red dashed arrow line.

Fig. 6. Comparison of SMGI and DGI with 2000 measurements under two-arm longitudinal deviation ΔZ=Z1−Z2 (μm) with fixed Z1=200  mm and variable Z2. (a) Reconstructions of the object “GI”; (b) Reconstructions of “zhong”; (c) PSNR curves corresponding to (a); (d) PSNR curves corresponding to (b).