High-Performance Fourier Iteration Ghost Imaging Under Low Sampling Rate

Ghost imaging (GI) has attracted the attention of researchers due to the characteristics of nonlocal imaging and strong anti-interference ability, but obtaining high-quality images needs a large number of samples, leading to the long imaging time. A Gerchberg Saxton-like (GS-like) algorithm is proposed for achieving Fourier iteration GI, which can obtain high-quality reconstructed results under low sampling rate and reduce imaging time significantly. In GS-like algorithm, a portion of the Fourier spectrum and the actual bucket detector values obtained through Fourier GI (FGI) are used as initial constraints ensuring obtaining high-quality image under low sampling rate. Then use the reconstructed image to obtain the phase image of the target object by Gerchberg Saxton (GS) algorithm. The proposed method is verified through the experiment, with the low sampling rate of 0.03 and 0.09 while peak signal-to-noise ratio (PSNR) and correlation coefficient (CC) are 18.8, 27.8 dB and 0.91, 0.94, respectively, proving to be superior to the traditional GI methods.