Fig. 1. Processing procedure of POH with the error diffusion method. (a) Object and its CAH. (b) The error diffusion mode for odd rows, and black arrow denotes the scanning direction. (c) The error diffusion mode for even rows. (d) The calculated POH and its reconstruction.

Fig. 2. Numerical reconstructions of the resolution test chart using different methods. (a) Object image. (b) Numerical reconstruction with the BERD algorithm. (c) Numerical reconstruction with the proposed PA-BERD algorithm.

Fig. 3. Reconstruction quality improvement of different object images by the PA-BERD algorithm. (a)–(e) are the reconstructed images with the BERD algorithm. (f)–(j) are the reconstructed images with the proposed PA-BERD algorithm. (k) is the PSNR comparison of (a)–(j).

Fig. 4. (a) and (c) show the relationship between the value of α and the PSNR of the reconstructed images of (b) “Peppers” and (d) “Airplane”, respectively. The blue and red dots represent the simulation results of the BERD algorithm and the PA-BERD algorithm, respectively.

Fig. 5. Numerical reconstructions at different distances using different methods. (a)–(d) are the reconstructed images with the BERD algorithm at the distances of 40, 60, 70, and 80 mm, respectively. (e)–(h) are the reconstructed images with the proposed PA-BERD algorithm at the distances of 40, 60, 70, and 80 mm, respectively. (i) is the PSNR comparison of (a)–(h).

Fig. 6. (a) and (b) are the schematic and photograph of the experimental setup, respectively. (c) shows the object image patterns “HOLOLAB” and “star”.

Fig. 7. Optical reconstruction results using different methods. (a) and (b) are the POHs and the corresponding optical reconstructed images and intensity distributions of “HOLOLAB” and “star” using the BERD algorithm, respectively. (c) and (d) are the POHs and the corresponding optical reconstructed images and intensity distributions of “HOLOLAB” and “star” by using the proposed PA-BERD algorithm, respectively.